Electric motor with multiple star points

Abstract

An electric motor comprises the following features: a rotor (100) with a first number of permanent magnets (101, 102, 103, 104), each permanent magnet comprising a first circular sector; a stator (200) with a second number of pole feet, each pole foot of the second number of pole feet having a coil wound around it, and each pole foot comprising a second circular sector smaller than the first circular sector, a first group of coils (301, 302, 303, 304) being electrically connected to one another via a first star point (311), a second group of coils (321, 322, 323, 324) being electrically connected to one another via a second star point (331), the second star point (331) being electrically insulated from the first star point (311), and a coil of the first group of coils being arranged between two coils of the second group of coils;and a control (500) for supplying the first group (301, 302, 303, 304) of coils with drive signals in order to provide the rotor (102) with a torque relative to the stator (200), and for supplying at least one coil (321, 322, 323, 324) of the second group of coils with a control signal that differs from the drive signals.

Description

[0001] The present invention relates to electric motors, and in particular to high-speed electric motors. These electric motors can be used universally. However, particular emphasis is placed on their use in a heat pump.

[0002] EP 2 549 113 A2 discloses a magnetic rotor and a rotary pump with a magnetic rotor. The rotor is magnetically driven and mounted within a pump housing inside a stator of the rotary pump for pumping a fluid without physical contact. The rotor is also encapsulated by an outer encapsulation containing a fluorinated hydrocarbon. Inside the encapsulation, the rotor comprises a permanent magnet encased in a metal shell. The rotary pump includes a pump housing with an inlet for supplying a fluid and an outlet for discharging the fluid. The fluid is, for example, a chemically aggressive acid containing a gas, such as sulfuric acid with ozone. For pumping the fluid, a magnetic rotor is magnetically mounted within the pump housing without physical contact. The rotor is further equipped with a magnetic drive comprising electrical coils.The stator is made of laminated iron, which is magnetically connected to the rotor's permanent magnet. The drive is a bearingless motor, in which the stator serves simultaneously as both the bearing and drive stator. The rotor is a disc rotor, with its axial height being less than or equal to half its diameter.

[0003] The ETH dissertation No. 12870, "The Bearingless Disc Motor," by N. Barletta, 1998, discloses magnetically levitated disc motors. Magnetic bearings operate completely without contact, wear, maintenance, or lubrication. Two controllable electromagnets, including electronic control, are required for the active stabilization of one degree of freedom. The bearingless disc motor is used within a bearingless blood pump as a bearingless disc motor with an active axial bearing, as a miniature disc motor, or as a bearingless bioreactor. By combining passive reluctance magnetic bearings and a bearingless motor, it is possible to fully support a disc rotor with only two actively stabilized radial degrees of freedom. The requirement for a large air gap, necessary in hermetic systems, is met by selecting a bearingless permanent magnet synchronous motor.A bearingless disc motor suitable for driving an axial pump for cardiac support is designed for speeds of 30,000 revolutions per minute, resulting in a smaller size.

[0004] Commercial electric disc motors are also known as "pancake motors." The motor concept described in the two preceding references is characterized by the fact that the stator extends around the rotor. Such motors are also called internal rotor motors.

[0005] The problem with the internal rotor design is that the stator must always be larger than the rotor, meaning that the size and design of the rotor are always limited by the stator housing, or rather, that the rotor dominates the design of the stator. This limits the application range of such a disc motor designed as an internal rotor.

[0006] Furthermore, disc motors inherently suffer from the problem that the rotor, regardless of whether it is designed as an internal or external rotor, is subjected to pressure differentials or pressures in certain directions. These pressures cause a bearing to be loaded in the direction of the pressure acting on the rotor, thus increasing wear. Alternatively, if rotor deflection is permitted, the rotor will deflect in this direction, necessitating the provision of clearance for this deflection. Particularly when the pump is used to transfer a medium from a pressure zone with a lower pressure to a pressure zone with a lower pressure, or even to generate such a pressure differential in the first place, complex design measures must be implemented to either achieve the required wear resistance or to provide sufficient clearance for the resulting deflection.

[0007] A disadvantage of electric motors, and especially those operated in warm environments or designed to deliver high power, is the pervasive heat generation. High temperatures within the electric motor negatively affect the permanent magnets typically located on the rotor. If the stator of a motor becomes too hot, the heat is transferred through the motor gap to the rotor and the permanent magnets there, causing all the associated problems. Furthermore, the heating of the stator itself is also critical. The stator is typically equipped with coils. Heating of the coils can lead to high thermal stress. This high thermal stress in the coils can, over time, lead to fatigue of the coil wire insulation. Additionally, problems can arise regarding delamination of the stator core, which is made of a sheet metal material.Furthermore, due to increased temperatures or high continuous thermal stress, deformations or warping in the stator can lead to the motor no longer running as smoothly as it should or could.

[0008] Especially in high-speed motors with rotational speeds exceeding 30,000 revolutions per minute, even if the pressure in the motor gap is lower than the ambient pressure, the friction with the gas present there is still so high that the rotor's permanent magnets, which are located directly in the motor gap, are exposed to this high frictional energy and the resulting heat. Permanent magnets have the property that their functionality / magnetization diminishes when they become too hot. In certain cases, this damage is even irreversible and can lead to the complete failure of the entire electric motor. However, even during operation, it is crucial for all parameters of the electric motor that the permanent magnets are kept within an optimal temperature range, which is by no means guaranteed due to the high heat generation caused by friction in the motor gap.

[0009] EP 2 975 731 A2 discloses a disc rotor for an electric machine with a circular or annular disc-like rotor body and permanent magnets arranged circumferentially adjacent to each other on the rotor body. In particular, the rotor body comprises a first material for dissipating heat in the radial direction and further comprises a second electrically non-conductive material in the area of ​​the permanent magnets. Furthermore, in order to reliably hold the permanent magnets on the support element, the support element is provided with a circumferential rim against which the permanent magnets can be supported externally. This rim, like the area in which the permanent magnets are inserted, is made of a thermally conductive material, such as aluminum.

[0010] In press release 107 / 2014 dated October 24, 2014, the Vienna University of Technology reports on electric motors that operate without error-prone sensors. Normally, in such electric motors with magnetic bearings, sensors measure the position of the magnet and transmit the data to control electronics that regulate the electromagnets. However, such sensors repeatedly present the same problems.

[0011] Sensor technology is costly, space-consuming, and always particularly prone to failure. Thin wires and delicate solder joints in sensors are prone to failure and are therefore responsible for many motor failures. The synchronous machine system developed at TU Wien takes a different approach. It utilizes the existing cables that supply power to the electromagnet. Short electrical test pulses are sent through these lines, and the current position of the rotor can be derived from the electrical response. The electrical pulses required for this last only a few millionths of a second. Appropriately designed electronics adjust to the motor's speed and control the electromagnets.

[0012] Press release 27 / 2015 from the Vienna University of Technology, dated March 20, 2015, describes floating rotors, for example in electric motors, using sensorless control of magnetic bearings. Contactless bearings are particularly important where friction losses must be minimized and where, as in vacuum pumps, abrasion must be absolutely avoided. Magnetic bearings are also used in drive shafts that need to reach particularly high speeds. With magnetic bearings, the rotor's position is electronically fixed. Electromagnets used to readjust the rotor's position also function as sensors. The rotor and the electromagnetic coil are coupled by the magnetic field. By measuring the change in the current in the coil over time, the rotor's position can be calculated.

[0013] A potential problem with such approaches is that additional signals are required to measure the rotor's position in the magnetic bearing. These ultrashort test pulses must be generated and evaluated.

[0014] From DE 10 2013 009 776 A1, a brushless three-phase motor is known to which two electrically independent controllers and a motor drive system are connected. The motor drive system comprises a brushless motor with coils divided into a group A and a group B, a first controller that applies three-phase voltages to each coil of group A, a second controller that applies three-phase voltages to each coil of group B, and a sensor section that outputs an electrical signal corresponding to a rotation angle of a shaft to the first and second controllers. The first and second controllers apply the three-phase voltages to the corresponding coils, regardless of whether the other controller applies the three-phase voltages to the corresponding coils. An induced voltage of each coil in both group A and group B forms the three phases, and the phase of the induced voltage of each coil of the same phase in each of the groups is the same.There is a phase shift between the induced voltage of each coil contained in group A and the induced voltage of each coil contained in group B.

[0015] The object of the present invention is to create a more efficient and reliable concept for an electric motor.

[0016] This problem is solved by an electric motor according to claim 1 or a method for manufacturing an electric motor according to claim 19 or a method for operating an electric motor according to claim 20.

[0017] An electric motor according to one embodiment comprises a rotor and a stator. The stator has pole feet, around which coils are wound. Each pole foot comprises a circular sector smaller than the circular sector encompassed by a permanent magnet. The coils wound around the pole feet are divided into two groups. A first group of coils is electrically connected to each other via a first star point. Furthermore, a second group of coils is electrically connected to each other via a second star point, the two star points being electrically insulated from each other.

[0018] Furthermore, a coil of the first group is arranged between two coils of the second group. A control system is provided to apply drive signals to the coils of the first group of coils in order to impart torque to the rotor relative to the stator. In addition, the control system is designed to apply a control signal, different from the drive signals, to at least one coil of the second group of coils.

[0019] The control of the coils in the second group can be completely independent of the control of the coils in the first group because the two groups of coils are separated due to their different star points. Applying a potential to a coil in the first group, for example, the group used to drive the motor at a specific time interval, has essentially no effect on a coil in the second group. This is because the two star points of the coil groups are separate, meaning they are electrically isolated from each other or, depending on the implementation, even galvanically isolated.Furthermore, since the arrangement and dimensioning of the pole feet around which the coils are wound are designed with respect to the permanent magnets such that a pole foot comprises a circular sector that is smaller than a circular sector of the permanent magnet, the case will arise that only one group of the two groups of coils, i.e., one star point group, is needed to drive the motor.

[0020] The second star point group, on the other hand, is available for any other motor operation. Typically, a group of coils that were the drive coils in the first time interval becomes the group of so-called "idling coils" in the second time interval, contributing little or nothing to generating torque in the electric motor. Therefore, there is considerable flexibility regarding the control of these "idling coils" during the time interval in which they are not driving the electric motor.

[0021] One way to control the idler coils, i.e., the second group of coils, when the first group of coils are the drive coils during this time interval, is through speed control. Specifically, the voltage signal required for torque can be applied to the second group of coils at a specific switch-on time during the first time interval. This switch-on time is freely adjustable, and any change in the switch-on time directly affects the current flowing in the coil when it is contributing to the drive.Since the two coil groups are galvanically isolated from each other, for example, due to their different star points, the voltage applied to the coils during the interval when they are not contributing to the drive is arbitrarily adjustable. However, it immediately contributes to the drive when the coils do contribute, i.e., when they come close to a gap between two permanent magnets. By pre-setting the current through a coil simply by the moment the voltage is applied to that coil during the time interval when it is only acting as a running coil, the coil is, in a sense, already primed to be optimally energized during the interval when it contributes to driving the motor.

[0022] An alternative way to control the idler coil is to create a position control system for a magnetic bearing. If the motor is operated as a contactless motor with a magnetic bearing, the idler coils can easily be used for position control of the magnetic bearing. By measuring the currents in the coils currently driving the motor, the rotor's position relative to the stator can be determined, allowing a control signal to be sent to the idler coils either within the same time interval or the next. Since the idler coils do not contribute to the drive because they have a different neutral point than the drive coils, they can be controlled with voltages / currents to achieve position control. This can be achieved, for example, by using two opposing stator coils with a positive and a negative reference potential, respectively.The rotor is provided with a positive potential and ground. This exerts a force on the rotor because, due to this bipolar control, an attraction is generated on one side between the moving coil and the permanent magnet opposite it, while on the other side a repulsion, or a comparatively small attraction, is generated between the moving coil and the permanent magnet opposite it. This generated force vector leads to a change in the rotor's position and a corresponding compensation for any positional deviation of the rotor in order to stabilize, i.e., electronically fix, the contactless electric motor.

[0023] Further possibilities and applications for the electrical control of the idler coils during the interval when the idler coils are not contributing to the motor's drive can also be employed and, if necessary, combined with speed control and position control. Speed ​​control and position control can also be combined during operation. For example, if it is determined that position control is not currently necessary because the rotor is running stably relative to the stator, the system can switch from position control mode to speed control mode to implement a requested speed increase or decrease. For various reasons, it may also be necessary to intentionally change the position from an optimal position, for example, to brake the motor or to perform other operations.All these measures can easily be carried out on the idler coils, since energizing or supplying voltages to the corresponding idler coils has no effect on the supply of drive signals to the drive coils of the corresponding other group of coils.

[0024] Preferred embodiments of the present invention are explained in detail below with reference to the accompanying drawings. These show: Fig. 1 a schematic top view of a motor with an external rotor with permanent magnets and an internal stator with coils; Fig. 2A a tabular representation of the use of different coil groups in different time intervals as drive coils or idler coils; Fig. 2B a tabular representation of the control and selection of the drive coils of Fig. 1; Fig. 3 a schematic representation of the coil groups arranged relative to each other with the different star points 1 and 2; Fig. 4 an electrical circuit for a coil connection; Fig. 5 a schematic representation of the two star point groups and a corresponding control system; Fig. 6 A representation of the complete control system for the four drive coils or the four idler coils in the [context missing]. Fig. 1 shown embodiment; Fig. 7 Time diagrams of the currents measured for individual coils to illustrate speed control via the switch-on time in the “neutral” interval; Fig. 8. Control of the drive coils with a single control potential U M ; Fig. 9. Control of the tracking coils to achieve position control in a corresponding direction; Fig. 10 diagrams of currents in coils for detecting rotor misalignment; Fig. 11 a flowchart to illustrate the steps carried out for attitude control in an exemplary embodiment; Fig. 12 an overview of an example of a maximum motor gap and the corresponding control for position control; Fig. 13 a tabular representation for recording the position of the rotor offset and the corresponding control for position regulation based on a temporal change in the coil currents; and Fig. 14 a schematic cross-section for a heat pump with an electric motor according to the invention, which has the two star points.

[0025] Fig. Figure 1 shows a top view of an electric motor. The electric motor includes a rotor 100, which is in Fig. 1 is arranged externally. This means that the in Fig. The example shown is an external rotor. However, the present invention is equally applicable to an internal rotor, in which the stator is arranged around the rotor and thus the rotor is located inside the stator. The present invention is described below using the external rotor as an example only. However, the invention is readily transferable and applicable to internal rotors.

[0026] The rotor 100 comprises a first number of permanent magnets 101, 102, 103, 104, each permanent magnet comprising a first circular sector. The circular sector at which in Fig. In the embodiment shown in 1 with four permanent magnets, the angle is 90° or a slightly smaller angle, respectively, after the permanent magnets have been arranged as shown in Fig. As shown in Figure 1, the magnets are spaced slightly apart and have different polarities, so that, for example, permanent magnet 102 has a north pole facing the motor gap, while permanent magnet 101 or 103 has its south pole facing the motor gap. Although in Fig. While Figure 1 shows a rotor with four permanent magnets, rotors with three or more than four permanent magnets can readily be used. It is preferred that each permanent magnet covers substantially the same circular sector, so that, for example, when using three permanent magnets, each permanent magnet would cover 120° or slightly less than 120°, whereas when using, for example, eight permanent magnets, each permanent magnet would cover approximately 45°.

[0027] The electric motor further comprises a stator 200 with a second number of pole feet, which are in Fig. The pole feet are shown with dashed lines and are designated by circled numbers ①, ②, ③, ④, ⑤, ⑥, ⑦, ⑧, where these circled numbers also denote the corresponding coils wound around the pole foot. A second number of pole feet exists. In the case of the Fig. In the example shown, the second number of pole feet is greater than the first number of permanent magnets on the rotor. Specifically, eight pole feet, and thus eight coils, are used, while four permanent magnets are employed. However, other ratios are also usable, provided the number of pole feet is greater than the number of permanent magnets. Accordingly, each pole foot, or each coil wound around the corresponding pole foot, encompasses a second circular sector that is smaller than the first circular sector encompassed by a permanent magnet.

[0028] In particular, a first group of coils A1, A2, A3, A4 is electrically connected to each other via a first star point, as shown in Fig. Figure 3 shows this first group of coils, designated 301, 302, 303, and 304. The first star point, SP1, is shown at 311. Furthermore, a second group of coils, L1, L2, L3, and L4, is shown in Figure 3. Fig. 3, also designated 321, 322, 323, 324, is connected to a second star point SP2, designated 331. The second star point 331 is electrically isolated from the first star point 311. Furthermore, as shown in Fig. 3 is shown, and as it also appears from Fig. As can be seen from Figure 1, a coil of the first group of coils is arranged between two coils of the second group of coils.

[0029] Furthermore, there is a 500 controller, which is in Fig. As shown in Figure 5, the control unit 500 is designed to supply the coils of the first group of coils, i.e., the drive coils, with drive signals in order to impart torque to the rotor relative to the stator. Furthermore, the control unit 500 includes a function for performing follow-up control, i.e., to supply at least one coil of the second group of coils, which does not contribute to driving the motor, with a control signal that differs from a drive signal. The control signal that can be applied to a coil of the second group of coils is, for example, a control signal for opening a switch to disconnect a reference potential that was applied, for example, in the previous time interval. Alternatively, the control signal can be the application of a control signal to a switch to ground a terminal of a coil that is not connected to the star point.Alternatively, the control signal can also be a control signal to a switch to apply a reference potential that was not previously applied to a coil. However, the control signal can also include a voltage across one or more coils, or a current through one coil, or currents through multiple coils, for both the drive coils and the idler coils.

[0030] Preferably, a reference potential is applied at a specific time for speed control. Based on this reference potential, and assuming a known current rise time at the beginning of the next interval, when the idler coil reverts to being a drive coil, a specific current flows through the coil. If the switch-on time is chosen earlier, the time it takes for the coil to revert to being a drive coil is longer, and the current will be higher to increase the speed. However, if the switch-on time is chosen later in the time interval when the coil is not the drive coil, the current will be lower, and the speed will gradually decrease.

[0031] In alternative embodiments, applying a control signal to at least one coil of the second group of coils constitutes position control. Here, a signal applied to a specific coil (e.g., one of the four coils in the second group) can control a switch such that a reference potential is applied to that coil, while a ground potential or a negative reference potential is applied to an opposite coil. This exerts a force on the rotor to change its position, thereby equalizing any misalignment of the rotor relative to the stator or any unevenly sized motor gap, or compensating for or eliminating the misalignment.

[0032] Fig. Figure 2A shows a table illustrating which coils are the drive coils and which coils are the idler coils for a given interval. In the Fig. In the state of rotor 100 shown with respect to stator 200, the first group of coils A1, A3, A2, A4, also labeled 1, 3, 5, 7, is the group of drive coils because each coil is located opposite a boundary between two oppositely polarized permanent magnets. In contrast, the second group of coils L2, L3, L4, L1, also labeled 2, 4, 6, 8, does not experience any change in the magnetic field. Therefore, this group of coils does not contribute to the drive. This applies for the interval k. For the time interval k+1, the rotor has reached a certain rotation, such that the boundary between two oppositely polarized permanent magnets is no longer opposite coil A2, but rather opposite coil L3. Thus, coil L3 becomes the drive coil, and coils A2 and A3 become the idler coils during this time interval. This is shown in the second row of the table. Fig. 2A is shown. If the rotor has then rotated further such that the transition between two oppositely polarized permanent magnets is again opposite the drive coil A3, for example, then the first group of coils is again the group of drive coils and the second group of coils are the idler coils. This is shown for the interval k+2. An interval therefore has, in the case of the Fig. In the embodiment shown in 2A, the time length is one quarter of the time length required for a complete revolution of the rotor around the stator.

[0033] Fig. Figure 2B shows a tabular summary of the control and selection of the drive coils. For example, in interval k-1, coils 3 and 7 are connected positively, and coils 1 and 5 are connected to a negative voltage or ground. In interval k, coils 4 and 8 are then connected positively, and coils 2 and 6 are connected to a negative voltage or ground. Then, in the following interval k+1, coils 5 and 1 are connected positively, and coils 7 and 3 are connected negatively or ground. In the next interval k+2, coils 6 and 2 are connected positively, and coils 8 and 4 are connected negatively. In interval k+3, coils 3 and 7 are connected positively, and coils 1 and 5 are connected negatively or ground. The last row of the table in Fig. Figure 2B shows the following windings, whose control signals can be chosen arbitrarily because, due to the different star points for the winding groups, they do not influence the other winding group. In the interval k+4, the situation is the same as in the interval k, and so on. This ensures that the rotor rotates relative to the stator, as is typical for such rotating field machines.

[0034] Fig. Figure 3 shows the arrangement of the coils of the different groups such that one coil of one group is always arranged between two coils of the other group. Fig. Figure 4 shows an embodiment of a switching element that can be used to control a coil. It can have a control potential 400, which is connected to U nThe circuit is designated as being connected between a positive terminal and a ground terminal. Furthermore, there is a first switch S1, designated 401, and a second switch S2, designated 402. In a preferred embodiment, switch S2 can also be bridged, for example, with a freewheeling diode 403. A corresponding freewheeling diode, which is located in Fig. As shown in point 4 with a dashed line, switch S1 can also be bridged, such that the reverse direction of the diode is as shown in Fig. 4 is shown with a dashed line.

[0035] The switching element that is in Fig. As shown in section 4, it is preferably located at each coil connection in Fig. 3 arranged, wherein each coil terminal of the first group preferably has its own reference potential and thus also its own ground potential, such that a separate reference potential and a separate ground potential are available for each star point group. This results, as described in Fig. Figure 5 schematically shows a first star point group 501 with four coils and eight switches and a second star point group 502 with four coils and four switches, all of which are controlled by the controller 500. In particular, the controller 500 is configured to supply all control signals for the corresponding switches S1, S2 for each coil connection.

[0036] Furthermore, especially when the idler coils are used for position control, each star point group is equipped with corresponding current sensors in order to always detect the currents in the coils of the drive group and, depending on the detected currents, to determine and transmit the position control signals, which in turn are control signals to corresponding switches, either in the same time interval or in the next time interval.

[0037] Fig. Figure 6 shows a complete circuit with coils 301, 302, 303, 304 and switch groups S11, S12 for the first coil 301, S21, S22 for the second coil 302, S31, S32 for the third coil 303 and S41, S42 for the fourth coil 304. In particular, it is preferred that only a single reference potential U M to use.

[0038] This reference potential is located between terminal 600 and ground terminal 601. At a first node 602, the following are connected: switch S41 for the fourth coil 304, switch S11 for the first coil 301, switch S21 for the second coil 302, and switch S31 for the third coil 303. For clarity, this node 602 is connected multiple times in Fig. Figure 6 shows the diagram. Furthermore, a second switch is installed at an intermediate node, designated 604, as shown, for example, for S42 at 604. The second switch, S32, is installed at intermediate node 605, between this node and ground potential 601. Additionally, switch S22 is installed at another intermediate node 606, between this node and ground. Furthermore, switch S12 is connected between an intermediate node 607 and ground potential 601. Thus, each coil has, at the terminal not connected to the star point SP, both a switch to the reference potential and another switch to ground potential, with the respective intermediate node, such as 604, 605, 606, or 607, being precisely this coil terminal not connected to the star point.Furthermore, preferably at least every second switch, such as S12, S22, S32, S42, includes a freewheeling diode 611, 612, 613 or 614, respectively. Additionally, a current sensor is provided for each coil, designated 621 for the first coil, 622 for the second coil, 623 for the third coil, and 624 for the fourth coil.

[0039] Each current sensor measures the current through the corresponding coil, i.e., I1, I2, I3, I4. A resistor of known value can be used as the current sensor, and the voltage drop across this resistor is then measured to determine the current. Alternatively, a Hall effect sensor or something similar can be used.

[0040] The in Fig. The circuit shown in Figure 6 is preferably implemented in exactly the same way for both the first star point group 501 and the second star point group 502, but preferably using different voltage sources to supply the reference voltage U. Mbetween terminals 600 and 601, preferably using galvanically isolated ground potentials, and wherein at least the two star points of the two circuits are electrically isolated from each other. Insulation resistances of, for example, 1 kΩ may be sufficient in certain embodiments. However, significantly higher insulation resistances in the range of greater than 10 MΩ or even 100 MΩ are preferred in order to achieve true galvanic isolation between the two star point groups, such that as little interference as possible occurs between the drive control by a first star point group and the follow-up control by the second star point group.

[0041] Fig. Figure 8 shows a preferred circuit of the first star point group for the drive. In particular, coils A1, A2, A3, A4 are connected as shown in Fig. Figure 8 shows that a voltage equal to the reference voltage U is applied to coils A1 and A3. M / 2 Furthermore, a voltage equal to the negative half of the reference voltage U is applied to the other two coils A4 and A2. M / 2 Furthermore, the star point SP1 lies on a potential U M / 2 regarding the mass.

[0042] With regard to switch control, the in Fig. 8. Example circuit shown for the drive for a specific interval, namely for the interval k+1 of Fig. 2B as follows. Switch S41 is open and switch S42 is closed. In contrast, switch S11 is closed. Fig. Switch S12 is open, while switch S12 is closed. Furthermore, switch S31 is closed. Fig. Switch S32 is closed while switch S32 is open. Furthermore, switch S22 is closed while switch S21 is open. This results in the voltage situation of interval k+1 of Fig. 2B to achieve drive control.

[0043] Fig. Figure 9, on the other hand, shows an exemplary circuit for a control in the direction R in the above. Fig. The schematic coordinate system shown in Figure 9 illustrates this. Control along the direction R represents control in the negative x- or positive x-direction and in the negative y- or positive y-direction, as shown by arrow 901. For this purpose, one of the two coils L1, L3 is connected to ground, while the other coil is connected to the reference potential U. M is connected. The other two coils, L2 and L4, are open when regulation is to occur along the 90° direction. If one then considers the circuit in... Fig. Considering point 6, this means that in order to connect coil L1 to the reference potential, switches S11 and S12 are open. Furthermore, switch S31 is open for coil L3, and switch S32 is closed. Thus, the total potential U M The coils L1 and L3 are activated, and if coils L2 and L4 are not activated simultaneously, a force is exerted in the direction of arrow R 901. To switch coils L2 and L4 accordingly, preferably all switches S41, S42 and S21, S22 are fully open, so that the intermediate nodes 604 and 602 are floating.

[0044] Alternatively, to achieve a control in the direction of arrow R'902, the two coils L2, L4 must be connected between the reference potential U Mand ground, while the two coils L1 and L3 are floating. To achieve this, switches S21 and S42 would need to be closed and switches S22 and S41 opened. The same applies to switches S12, S11, S32, and S31, which would also need to be opened to bring coils L1 and L3 into the floating state.

[0045] If, on the other hand, it is preferred to achieve control along a direction other than direction 901, 902, then it is preferred to apply the voltage to both L1 and L2 and to connect L3 and L4 to ground. Then a force would be exerted on the rotor, which would be measured along the x- and y-axes of the diagram. Fig. 9 is. Since each direction can be set by a corresponding combination of two orthogonal directions in the sense of a parallelogram of forces, depending on the implementation, control in each direction is possible qualitatively but also even quantitatively.

[0046] For example, to connect coils L1 and L2 to the reference potential, switches S11 and S21 would be closed and switches S12 and S22 would be opened. Conversely, switches S41 and S31 would be opened and switches S42 and S32 would be closed.

[0047] The following will be based on Fig. Figure 7 shows a first embodiment for controlling the speed of the motor via the switch-on time. The intervals are shown in Fig. 7 is shown in Fig. 700. Furthermore, the respective ordinate axis in the time diagrams shows Fig. 7 each shows the current through a coil. In particular, the current waveform of a coil is also shown, which in the first interval k is in driving mode, i.e., belongs to the first star point group, but which in the second interval k+1 belongs to the second star point group and is therefore a moving coil. Furthermore, in Fig. Figure 7 in the top diagram shows a situation where the dashed line represents the current waveform of a different coil, each in the other group. This shows that in interval k, the solid current waveform represents the current waveform of a drive coil, and the dashed current waveform represents the decreasing current waveform of a running coil. Since the running coil has no influence on the drive, it is not a problem that the current drops slightly in interval k. The current could just as easily drop completely, as will be discussed later with reference to... Fig. 10 is shown. However, if the current drops slightly, this is achieved, for example, by using certain switch positions, as shown by Fig. The values ​​shown in Figure 6 are set. In any case, each coil is provided with a corresponding inductance L. Furthermore, the relationship between voltage and current across a coil is such that the voltage across the coil is equal to the product of the coil's inductance and the derivative of the current through the coil with time. Therefore, if a coil is acting as a moving coil, then when a voltage is applied at a specific time 705, the current will increase constantly, as shown by the dotted line in Figure 6. Fig. Figure 7 shows this with a constant rate of increase, since the inductance of the coil is constant. Therefore, at time 705, shortly before the start of interval k, the switch is positioned accordingly, as shown in Figure 7. Fig. 6 has been set out, to which in Fig. In the first time diagram, a voltage is applied to coil 7, whereupon the current increases until it reaches its value I1.

[0048] The point in time at which the current value I1 is reached determines the beginning of interval k, meaning that the coil under consideration changes from a moving coil to a driving coil. Due to this change, mutual induction occurs, and the current I1 does not increase further but remains constant as long as this coil contributes to the driving signal. Then, at the end of interval k, the coil no longer contributes to the driving signal but reverts to a moving coil, and the current through the coil in this interval is not particularly significant; it can therefore, for example, decrease slightly, as illustrated by the solid line in interval k+1.

[0049] The situation in Fig. Condition 7 is reached when the time interval Δt1 between the switch-on time 705 and the beginning of the interval k or k+1 has a corresponding value. This results in a specific current I1 of a certain magnitude. However, if, as in Fig. As shown in the second diagram, if the switch-on time 706 is chosen, which is earlier than the switch-on time 705, the switch-on time interval Δt2 becomes larger. Due to the constant increase in current in a coil when the coil is a moving coil, a maximum current I2 is now reached. This maximum current I2 does not increase further once the rotor has moved to the point where the coil is no longer a moving coil but a driving coil contributing to the drive.

[0050] The current I2 remains constant until the transition from interval k to interval k+1 occurs. At this point, the circuit is switched on by means of one or more control signals. Fig. 6 is controlled in order to disconnect the reference potential in some way, so that the current does not continue to rise, but instead falls, as described in Fig. Figure 7 shows that, simultaneously, another coil, which was the running coil during interval k, is switched on again with the larger time interval Δt2 to achieve the current I2 when interval k+1 begins, and so on. It follows that the switch-on interval Δt2, which is chosen to be larger than Δt1, results in a larger current in the coil when it becomes the driving coil. Since the current is proportional to the rotational speed, a higher rotational speed is achieved, so that the rotational speed Δt2 is greater for a larger time interval Δt2 than the rotational speed Δt1 when the time interval Δt1 is smaller.

[0051] Similarly, a reduction in speed can also be achieved if the switch-on time 707 is shifted closer to the interval limit than the switch-on time 705. This reduces the current in the coil, when it becomes the drive coil, to a value I3, which is smaller than I1. This results in a speed lower than the speed ΔH1, which is determined in the situation shown in the diagram above. Fig. 7 is reached.

[0052] This shows that speed control can be achieved simply by selecting the switch-on point, taking into account the linear increase in current due to the constant inductance of the coil and the constant applied voltage, while all switching operations are carried out on the idler coils. This prevents any instability from being introduced into the motor's operation, because the motor runs particularly smoothly due to the fact that the drive coils are already "prepared" for the correct state during the relevant time intervals when they are still idler coils.The electric motor therefore sees no increases or decreases in current through the coils that would cause uneven running; instead, the electric motor sees only constant currents from one interval to the next, but always alternating between a coil of the first star point group and a coil of the second star point group.

[0053] Alternatively or additionally, the fact that there are two different star point groups, and thus drive coils and idler coils, can also be used for position control of a magnetic bearing. This is shown schematically in Fig. 10 shown. Fig. Figure 10 again shows intervals k, k+1, k+2, where the solid line represents the current flow of a coil in the second star point group and the dashed line represents the current flow of a coil in the first star point group. In the Fig. In the uppermost current waveform shown in Figure 10, the situation is such that the rotor is perfectly aligned with respect to the stator and the motor gap for the coil under consideration is constant over a complete revolution. Therefore, the current I in a coil is constant when the coil is the drive coil. A constant current indicates that there is no offset. Then, when the drive coil becomes the running coil, as for example in the interval k+1 for the coil of the second star point group SP2, the current can also be, for example, different from that shown in Figure 10. Fig. In the embodiment shown in section 7, the voltage can be completely reduced to zero. This could be achieved, for example, by removing the connection of the coil, i.e., the intermediate nodes, for example 604, 605, 606, 607. Fig. 6 of the respective coil is grounded by the corresponding switch S42, S32, S22, S12. Then the coil would discharge and the current would drop, as described in Fig. 10 is shown. It can then be switched on again at a corresponding switch-on time, whereby this switch-on time can be chosen, for example, as shown in Fig. As explained in section 7, this applies when specific increases or decreases in speed are required. However, if the speed is to remain constant, the switch-on point would always be chosen at the corresponding interval Δt before the interval limit.

[0054] However, if a current sensor were to produce a curve like the one in the second illustration, Fig. If the measurement is taken at interval 10, where the current increases within the interval, this means that the motor gap "sees" by the coil in question increases as the rotor rotates around the coil, indicating a "positive offset." This means that the rotor and stator do not share exactly the same central axis, but rather that the central axes of the two motor components are offset. In particular, as already mentioned, the increasing current indicates a widening gap.

[0055] However, as it is in Fig. As shown in Figure 10, a falling current over the interval leads to a decreasing gap that the coil under consideration “sees” past the coil as the rotor rotates, which in turn means a negative offset.

[0056] The evaluation of these currents is preferably used to determine whether an offset exists at all, and in which direction the offset exists, in order to generate a corresponding position control by appropriately controlling the tracking coils in the same interval in which the measurement takes place, or in the next interval, i.e. in the interval that follows the interval of the measurement.

[0057] This refers to Fig. 13 Referenced. Fig. Figure 13 shows a table with different gap positions, in particular a position of maximum gap and a minimum gap. Furthermore, for the purpose of illustrating the nomenclature, the drive coils 1, 3, 5, 7 are shown in Figure 13. Fig. 13 shown, which correspond to the drive coils of the same numbers in Fig. 1 approximately correspond. In addition, currents i L1 , i L3 , i L5 and i L7represented by the coils. Furthermore, the table contains the derivatives of these currents D1, D3, D5, D7 with respect to time, specifically regarding their qualitative characteristics: whether the change in current is positive, indicating a widening gap; whether the change is negative, indicating a narrowing gap; whether the overall change is zero, indicating optimal alignment; or whether, in this interval, both an increase and a decrease in current occur, meaning the average derivative is zero or very small.

[0058] It should be noted that the derivative can be calculated arbitrarily, as long as a current or average change in current over time is calculated. The derivative could therefore, for example, also be calculated simply by taking the difference between two current values ​​in the interval. Depending on the implementation, this is sufficient because the corresponding evaluation of how the position control is performed is already carried out by the [missing information]. Fig. The embodiment shown in 13 is qualitative.

[0059] For example, if it is determined that derivatives D1 and D3 are positive and derivatives D5 and D7 are negative, then a difference between D1 and D5 would result in a large positive value, and a difference between D3 and D7 would also result in a large value. If these two combined values—that is, the difference between D1 and D5 on the one hand, and the difference between D3 and D7 on the other—both yield large values, this means that the maximum gap lies between coils 3 and 4 or 3 and 5, and the minimum gap lies between coils 7 and 8 or 7 and 1. Fig. 1. If, on the other hand, the maximum gap were to lie, for example, between 7 and 8 and 3 and 4, then the derivatives D1, D3 would be negative and the derivatives D5, D7 positive, so that the two combined values ​​D1-D5 on the one hand and D3-D7 on the other would lead to large negative values, as shown in column 1300 of Fig. Figure 13 illustrates this. Accordingly, an evaluation would also determine whether the maximum gap, and therefore also the minimum gap, is located at a position as shown in one of the last four columns. For example, if the maximum gap were opposite coil 7, as shown in Figure 1302, the minimum gap would be opposite coil 3. D1 would be negative, D3 would be nearly zero, D5 would be positive, and D7 would be nearly zero. D1-D5 would then result in a large negative value, and D3-D7 in a value nearly zero. Therefore, due to the different evaluations of the combined values ​​D1-D5 on the one hand and D3-D7 on the other, which can be large positive, large negative, or zero, different localizations of the maximum and minimum gaps can be concluded, as shown in the first two rows in Figure 1302. Fig. 13 is shown.

[0060] The last two lines in Fig. Line 13 again shows which coils can be controlled for position control. A distinction is made between whether the position control is performed in the same interval or in the next interval, i.e., when the roles of the drive coils and the idler coils are reversed. Depending on the implementation, if the intervals are large enough, the position control can be performed in the same interval. For the first line, 1301, where the maximum gap is between coils 3 and 4 or 3 and 5, i.e., somewhere opposite coil 4, a positive potential could easily be applied to coil 4 and a negative potential, or ground potential, to coil 8 for position control. This would be done with a polarity that causes the rotor to be attracted towards coil 4, thus significantly reducing the motor gap.If, on the other hand, position control were not carried out in the same interval, but in the next interval, then with a direction of rotation such as that in 111 in . Fig. As shown in 1, coil 3 or coil 5 can be used for position control, for example by applying a positive potential, since coils 3 and 5 in the interval k-1 are then no longer the drive coils, but the position control coils.

[0061] Similarly, for example, as shown in column 1302, one would in the same interval in which the maximum gap of coil 7 in Fig. Position control is performed when coil 1 is opposite, using position control coils 6 and 8 to apply, for example, a positive potential and the other coils 2 and 4 to apply a negative potential or ground. However, if position control is performed in the next interval, coil 7 itself can be used to apply a positive signal and a ground potential or a negative potential to the opposite coil 3.

[0062] This is exemplified in Fig. 12 compiled. Is the gap, for example, at its maximum after coil 3, i.e., between 3 and 4 or between 3 and 5, as shown in the first column in Fig. As shown in Figure 13, the drive coils are coils 1, 3, 5, and 7. This means that, due to the gap at the specified position, the current in coils 1 and 3 increases and in coils 5 and 7 decreases. To achieve position control, if the position control is performed at the same interval, the tracking coil 4 would be connected to ground and the tracking coil 8 to U. M They can be connected, or vice versa. Alternatively, the tracking coil 4 could be connected to ground and the tracking coil 8 to U. M It could be, or it could be that the tracking coil 8 is connected to ground and the tracking coil 4 to U. M can be used. The appropriate alternative for a motor depends on the winding direction of the coils and the orientation or magnetization of the permanent magnets.

[0063] In general, one embodiment consists of the position control being designed such that, at maximum gap, the polarity is selected as it was in the last drive interval. For example, if coil 4 was positive (with U) in the last drive interval. M ) was connected, coil 4 for position control is also positively (with U M ) connected. This results in an attraction and a reduction of the gap relative to coil 3. However, if coil 4 was negative (with -U) in the last drive interval M or ground) connected, coil 4 for position control is also negatively charged (with -U). M ) or ground). This would also achieve an attraction and reduction of the gap at 3. The opposite coil is then connected in the opposite direction, i.e., with respect to coil 8.

[0064] Fig. Figure 11 shows a flowchart for the implementation of various measures, which are carried out by control 500, for example, which is in Fig. As shown in section 5, the following steps will be carried out.

[0065] First, the interval k is considered. Within interval k, currents in the drive coils are preferably measured, as shown in Figure 1100. For this purpose, corresponding current waveforms are created, for example, as shown in [reference missing]. Fig. 7 or Fig. Figure 10 is shown. Generally, however, it is preferred to record a current profile over time, although a first current value at a specific point in time and a second current value, e.g., at the midpoint of the interval, would also suffice. In step 1102, the currents are then derived to obtain, for example, the values ​​D1 to D7, which are shown in Fig. 13 have been shown.

[0066] Then, in step 1104, the derivatives are combined, preferably for coils 1, 3 on the one hand and 2, 4 on the other, i.e., for opposing coils according to the nomenclature as described in Fig. 3 is shown. Opposing coils are considered in their entirety with regard to the nomenclature as described in Fig. As shown in Figure 13, coils 1 and 5 and 3 and 7 are shown.

[0067] In step 1106, a detection of a rotor positional offset is performed to determine the direction of the offset. This is done, for example, by evaluating lines 1310 and 1311 of Fig. 13 instead, in order to be able to deduce the first two lines or to carry out a corresponding control of the tracking coils in order to reduce the positional offset, as can be seen from the last two lines in Fig. 13 is shown. For example, if row 1310 has a large positive value and row 1311 has a large negative value, this would be shown in the fifth row of Fig. 13. This results in coils 2, 6 being activated for position control if the position control is carried out in the same interval, or coils 1, 2 being activated if the position control takes place in the next interval.

[0068] The control of the flyback coils is at 1108 in Fig. Figure 11 illustrates this. If the intervals are large enough, or if the measurement over only a fraction of the interval is sufficient to determine the rotor offset with reasonable certainty, then the drive coils can be activated to reduce the positional offset, as shown in Figure 1108. Fig. As shown in Figure 11, the action will take place in the same interval. However, if the entire interval is used to perform steps 1100 to 1106, the control of the tracking coils will occur in the next interval k+1.

[0069] Furthermore, another measurement takes place in the next interval, as was the case at 1110. Fig. 11 is shown. Accordingly, after the interval k-1, the control of the tracking coils is carried out in the interval k based on the positional offset from the interval k-1, as is the case with 1120 in Fig. Figure 11 illustrates this. Therefore, if position control occurs in the next interval, the tracking coils of the next interval will be activated due to the positional offset from the previous interval.

[0070] Although the present invention is described by reference to the various figures and in particular to Fig. As shown in Figure 1 with four permanent magnets and eight coils, it should be noted that any other combination of magnets and coils can also be used, as long as the number of permanent magnets is less than the number of coils. Thus, there is always a first group of coils, electrically connected to each other via a first star point, which are used for drive control, while a second group of coils, electrically connected to each other via a second star point, are the idler coils. The number of coils in the drive group and the idler group can be 2, but can also be 3 or, preferably, 4, although even larger numbers of coils are also possible.

[0071] In exemplary embodiments, the procedure, particularly for position control, is based on the fact that the current through a coil increases, with a constant applied voltage, when the gap of the motor opposite the coil becomes larger. For speed control, for example, the principle is that the current for the drive coils can be switched on and off in the preceding time interval, in which the drive coils are still running coils. This allows the required current to be maintained for a drive coil from the beginning to the end of the time interval, since the current can then be switched on or off in the preceding or subsequent time interval, where the coil is only the center running coil, without affecting the motor's behavior.Therefore, it is also irrelevant whether, in the tracking interval, i.e., in the interval in which a coil is the tracking coil, the current through the coil is completely switched off, i.e., switched to zero, or left at a certain level.

[0072] In particular, to switch off the current, in an embodiment of the present invention, switches such as those used in Fig. 4 or Fig. As shown in Figure 6, the corresponding coil is disconnected from the potential, for example, by switch S41 for coil 304, while switch S42 is simultaneously closed. Alternatively, switch S42 could also be left open. Nevertheless, current equalization and a corresponding discharge of the coils take place, particularly via the freewheeling diodes, which ensure that potentials do not fluctuate excessively but are kept within acceptable limits by the functionality of the freewheeling diodes. It is also particularly advantageous that by switching off the coil during the interval in which it is not contributing to the drive, no current needs to be conducted through a resistor and therefore is not converted into heat. Instead, a current flows between the coils on the one hand and the power supply on the other, and in particular a smoothing capacitor in the power supply.This ensures that energy stored in the coils, which is no longer needed when the coil is in standby mode, is transferred to the power supply and supplied from there in the next cycle. This means there is no energy loss, but rather a simple energy transfer between the motor and the power supply, resulting in significantly reduced cooling requirements and the lowest possible power consumption.

[0073] Regarding position control, it should also be noted that this is qualitative or preferably even quantitative. Proportional control is particularly preferred. For this purpose, the coils, as used, for example, in… Fig. The coils, as determined in the last two lines (13), are supplied with a specific current. However, it is preferred to supply the coils with the same reference voltage U. Mto apply to position control, and accordingly only the relevant switches, as they are in Fig. 6, for example, to control. For this purpose, it is preferred to apply corresponding time profiles to the switches so that they are ultimately controlled depending on a specific measured offset of the quantitative values ​​determined in lines 1310 and 1311.

[0074] Depending on the corresponding rotational speed, intervals in the microsecond range will suffice. For example, if a required rotational speed is 2000 revolutions per second, which corresponds to approximately 120,000 revolutions per minute, then one rotational cycle is 62.5 µs long. In such a cycle, which is divided into, for example, four intervals, each approximately 15 µs long, it is sufficient for small to medium deviations to use the reference potential U for 1 to 3 µs in the tracking interval. M from Fig. 6 with appropriate switch settings to achieve position control.

[0075] Furthermore, it should be noted that position control is particularly suitable for magnetically levitated motors that do not have contact bearings, such as ball or roller bearings. Such magnetically levitated motors are preferably designed as disc motors, which are stable along the axial direction of the axis of rotation but must be controlled along the radial direction, i.e., with respect to the radial offset of the rotor relative to the stator. However, speed control can also be implemented with motors equipped with contact bearings; a magnetic bearing is not absolutely necessary to perform speed control in conjunction with two coil groups with different star points and different drive groups.

[0076] Fig. Figure 14 shows a preferred application of the disc rotor motor using the example of a heat pump. The heat pump comprises an evaporator 300, a compressor 420, and a condenser 509, wherein the compressor 420 includes the electric disc rotor motor, which, with reference to the Fig. 1a to Fig. 5 has been described.

[0077] In addition to the elements of the disc rotor motor, which has been illustrated, for example, with reference to one of the preceding figures, the compressor further comprises a guide chamber 410, which is arranged radially in order to convey the working steam supplied by the moving element 105, which has been drawn in by the evaporator 300, and ultimately to increase the pressure to the required pressure in the condensation zone 510 in the condenser 509.

[0078] The liquid to be cooled flows into the evaporator via an evaporator inlet 340. The cooled working liquid flows out of the evaporator via an evaporator outlet 360. To ensure that the radial wheel 105 draws in only steam and not water droplets in addition to the steam, a droplet separator 306 is also provided. Due to the low pressure in the evaporator 300, some of the working liquid supplied to the evaporator 300 via the evaporator inlet 340 is evaporated and drawn through the droplet separator 306 over the second side 105b of the radial wheel 105, conveyed upwards, and then discharged into the guide chamber 510. Compressed working steam is introduced from the guide chamber 510 into the condensation zone 510. The condensation zone 510 is further supplied with working fluid to be heated via a condenser inlet 512, which is heated by condensation with the heated steam and is discharged via a condenser outlet 514.Preferably, the condenser is designed as a "shower" condenser, so that a distribution of liquid in the condensation zone 510 is achieved via a distribution device 516. This ensures that the compressed working steam is condensed as efficiently as possible and the heat contained in it is transferred to the liquid in the condenser.

[0079] At the in Fig. In the embodiment shown in Figure 14, a motor housing 110 is also shown, which simultaneously forms the upper housing part of the condenser or liquefier 509. Furthermore, a connecting line 80 for the coils of the stator 200 is connected to a control unit 600 in order to carry out the corresponding speed controls and simultaneously also the active bearing via a preferably used magnetic bearing, as shown in Figure 14. Fig. 7 has been described. The controller also provides the functions of the radial detection 270 and the radial control / regulation 280.

[0080] Furthermore, in Fig. Figure 14 shows an implementation in which the disc rotor motor has a solidly encapsulated block 630 made of potting material, which is sealed against the motor housing 110 by a sealing ring 603, so that a pressure-tight separation between the exterior and interior takes place. Both the coil holder and the coils are surrounded by an encapsulation material which is in Fig. 14 is shown as being formed integrally with the solid block 630. However, this need not be the case. It is preferred, however, to bring about a separation by means of the encapsulation material extending in the motor gap, such that the coils are not arranged in the low-pressure area present within the motor housing.

[0081] Furthermore, in the Fig. In the embodiment shown in Figure 14, the stator is arranged in a recess defined by an upper side 105a. However, in other embodiments, the rotor can be designed without a recess, so that the area consisting of magnet 101, return element 202, and bandage 203, as shown in Figure 14, is not visible. Fig. As shown in 14, it is mounted on a radial wheel with a flat top.

[0082] Out of Fig. 14 further shows that the moving element connected to the rotor 10 is the radial wheel or paddle wheel 105, which is intended to compress and thus heat the working steam supplied by the evaporator in conjunction with the guideway 410, so that heat is pumped from the evaporator to the condenser.

[0083] Preferably, the rotor is mounted relative to the stator by a magnetic bearing and is held axially relative to the stator by the permanent magnets on the rotor side and the electrical coils on the stator side, without any specific control mechanism. A radial sensing device and a radial control / regulating device (500) are provided. The radial sensing device detects the position of the rotor relative to the stator, and vice versa. The result of the radial detection is communicated to a radial control / regulating device via a sensor line. This device generates the corresponding actuator signals via actuator signal lines at the rotor or the stator, depending on the implementation. The coils are then controlled to position the rotor relative to the stator based on the actuator signal, such that the motor gap around the entire rotor is of a similar size and the rotor does not touch the stator.

[0084] In one embodiment, the rotor is located inside and the stator is located outside. This is therefore an internal rotor, in contrast to, for example, Fig. 1.. In principle, however, the magnetic bearing is similar in both cases, using the example of a reluctance bearing, in that axial control does not take place, while radial control is carried out by the radial sensing device and the radial control / regulation device.

[0085] Although certain elements are described as device elements, it should be noted that this description can equally be seen as a description of steps in a process and vice versa.

[0086] Furthermore, it should be noted that a control mechanism, for example one effected by element 500, can be implemented as software or hardware. The control mechanism can be implemented on a non-volatile storage medium, a digital or other storage medium, in particular a floppy disk or CD with electronically readable control signals that can interact with a programmable computer system to execute the corresponding method for operating a heat pump. In general, the invention thus also includes a computer program product with program code stored on a machine-readable medium for carrying out the method, when the computer program product runs on a computer. In other words, the invention can also be realized as a computer program with program code for carrying out the method when the computer program runs on a computer. Reference symbol list 80 connection cable 100 Rotor 101, 102, 103, 104 permanent magnets 105 Area to be rotated (e.g. radial wheel) 105a upper side of the radial wheel 105a lower side of the radial wheel 110 engine housings 200 Stator 202 magnetic return element 203 Bandage 300 evaporators 301, 302, 303, 304 Drive coils 306 droplet separators 321, 322, 323, 324 second group of coils 311 first star point SP1 331 second star point SP2 340 Evaporator inlet 360 evaporator drain 400 Reference potential U n 401 first switch 402 second switch 410 Control Room 420 Compressor 500 control 501 first star point group 502 second star point group 509 Condenser 510 Condensation zone 512 Condenser inlet 514 Condenser outlet 516 Condenser distributor 600 reference nodes 601 mass nodes 602 Reference potential connection node 603 Sealing ring 604, 605, 606, 607 Intermediate nodes 611, 612, 613, 614 freewheeling diodes 621, 622, 623, 624 Current sensors 630 Stator block 700 interval 705 Switch-on time 706 Switch-on time 707 Switch-on time 901 first attitude control direction 902 second attitude control direction 1100 Measurements of currents 1102 Diverting the currents 1104 Combining the derivatives 1106 Detect 1108 Controlling the tracking coils 1110 Measuring in the interval k+1 1120 Control in interval k 1300, 1301, 1302 columns of Fig. 13 1310, 1311 lines of Fig. 13

Images