DE102020117796A1 - Method and device for adjusting a holding current of an electrical machine - Google Patents

Abstract

The invention relates to a method for operating an at least three-phase electronically commutated electrical machine (2) with a stator winding (3) with a plurality of phase strands (4) and a rotor (5), the electrical machine (2) being pulse-width modulated with block commutation by applying generated control potentials (V1, V2) at phase connections (7) for energizing phase strands (4), with the following steps: - detecting (S5) at least one measurement potential (M1, M2) that changes after a measurement pulse is applied to one of the phase connections (7) sets;- when the electrical machine (2) is at a standstill, holding (S6) the rotor (5) against a load torque using a holding position control, the control variable of which depends on the at least one measurement potential (M1, M2) or corresponds to it and which specifies the control potentials (V1, V2) of the block commutation as the manipulated variable.

Description

Technical background

The invention relates to brushless electrical machines, in particular methods for operating electrical machines. The invention also relates to measures for reducing a holding current for electronically commutated electrical machines when holding against a load moment.

Technical background

Brushless electric machines have stator coils to generate an excitation magnetic field. Depending on the position of the rotor, the excitation magnetic field is set in such a way that a moment is exerted on the rotor and the rotor moves as a result.

Electrical machines of this type can have star-connected or polygon-connected stator arrangements. Depending on the area of application and the operating requirements, the electrical machines are operated with a type of commutation that generates a stator magnetic field. Block commutation represents a simple type of commutation in order to energize the phase strands of the stator winding of such an electrical machine. With block commutation, the stator coils of the stator arrangement are driven with a commutation pattern, with one of the phase connections being connected to a high supply potential and another of the phase connections being connected to a low supply potential, while the one or more other phase connections remain without current.

Knowledge of the rotor position is necessary for the operation of a brushless electrical machine in order to switch over the commutation pattern present at the phase connections in such a way that the electrical machine can be operated and a rectified torque is generated.

The operation of an electrical machine using block commutation makes sense when the rotor position is detected without sensors. Information about the rotor position can be determined by measuring a potential at a non-energized phase connection, the measured potential being determined, among other things, by an inductance of the energized stator windings that is dependent on the rotor position. In this way, information about the rotor position can be assigned to the measured potential.

Electrical machines of this type can be used in applications in which, after the electrical machine has stopped, a holding torque must be applied against an acting load torque. In particular when using block commutation, depending on the rotor position, high phase currents may be required to hold the rotor in a holding position in order to hold the rotor against a load torque, particularly in rotor positions in which a rotor pole is directly aligned with a stator tooth.

Furthermore, due to the reduction gear, which is generally coupled to the drive shaft, a larger tolerance range can be permissible for the holding position.

It is the object of the present invention to provide an improved method for setting a holding current for holding an electrical machine in a holding state, with the holding current in particular being minimized. Furthermore, it is an object of the present invention, when holding the rotor of the electrical machine, to essentially avoid movement of the rotor when the load torque changes.

Disclosure of Invention

This object is achieved by the method for setting a holding current for holding a rotor on the electrical machine in a holding position according to claim 1 and a corresponding device and a motor system according to the independent claims.

Further developments are specified in the dependent claims.

• - Erfassen mindestens eines Messpotenzials, das sich nach Anlegen eines Messimpulses an einen der Phasenanschlüsse einstellt;
• - bei einem Stillstand der elektrischen Maschine, Halten des Läufers gegen ein Lastmoment mithilfe einer Haltestromregelung, deren Regelgröße von dem mindestens einen Messpotenzial abhängt oder diesem entspricht und die als Stellgröße die Ansteuerpotenziale der Blockkommutierung vorgibt.

According to a first aspect, a method for operating an at least three-phase electronically commutated electrical machine with a stator winding having a plurality of phase strands and a rotor, the electrical machine being operated with block commutation by applying pulse-width-modulated control potentials to phase connections for energizing phase strands, with the following steps:

• - Detecting at least one measurement potential that occurs after applying a measurement pulse to one of the phase connections;
• - When the electrical machine is at a standstill, the rotor is held against a load torque using a holding current control, the control variable of which depends on the at least one measurement potential or corresponds to it and which specifies the control potentials of the block commutation as the manipulated variable.

Electronically commutated electrical machines can be operated with commutation, with stator coils of a stator winding being energized alternately depending on a rotor position. The simplest type of commutation is the so-called block commutation, in which a stator magnetic field is set, the orientation of which is based on the positions of the energized stator coils. Depending on a rotor position, the stator coils are energized in such a way that the alignment of the stator magnetic field leads the rotor magnetic field in the direction of movement of the rotor.

With block commutation, a supply voltage is applied between two of the phase terminals of the electrical machine, depending on the rotor position, regardless of whether the brushless electrical machine is configured in a star connection or a polygon connection. The selected phase connections depend on the current rotor position. The strength of the stator magnetic field and thus the motor torque acting on the rotor can be set by specifying the phase voltages applied to the phase connections. The phase voltages can be set using a pulse width modulation method. For reasons of cost, an electrical machine can be designed without sensors. In sensorless operation, a rotor position-dependent inductance of the stator winding is determined by measuring an electrical variable and assigned to a rotor position.

A typical inductance measurement is performed with block commutation by measuring and evaluating a measurement potential at a phase connection of the electrical machine with no current, whose potential is essentially dependent on the inductance of the stator winding. In order to determine an indication of the inductance of the stator winding, which depends on the rotor position, a measurement pulse is applied to the energized phase connections and a resulting measurement potential is measured at a non-energized phase connection. Provision can be made for the measurement pulse to be formed as part of a pulse-width-modulated generation of the control potential, or to be superimposed on the pulse-width-modulated generation of the control potential. Alternatively, the commutation of the electrical machine can be briefly interrupted in order to apply the measuring pulse and carry out the measurement.

Furthermore, two measurement potentials can be detected, in particular temporally within a pulse width modulation cycle of the pulse width modulation method, depending on the application of two opposing measurement pulses, with a measurement potential difference (R) being determined from the two measurement potentials, with the holding current control receiving the measurement potential difference (R) as a control variable.

Thus, in order to avoid the measuring potential being falsified by an induction voltage due to the action of a counter-electromotive force (back-EMF) when the rotor moves, a positive measuring pulse and a negative measuring pulse can be applied in close succession to the current path through the Stamped stator winding and the corresponding measurement potentials are measured at the relevant non-energized phase connection. By forming the difference between the measurement potentials obtained in this way, information on the inductance of the stator winding can be obtained, with the influence of the back EMF being eliminated when the rotor moves.

As a result, information on the rotor position with regard to an electrical rotor position angle (rotor angle) can be determined using the determination of the measurement potential difference. The electrical rotor position corresponds to the mechanical rotor position multiplied by a pole pair number of the rotor poles.

Essentially, a measurement potential difference of 0 volts corresponds to an alignment of one of the rotor poles in the direction of an energized stator coil, which corresponds to an electrical position angle of 0°. If there is an offset between the rotor pole and the stator tooth, the measurement potential difference changes accordingly, since the inductance of the stator winding changes. When the rotor rotates, the measurement potential difference is essentially sinusoidal in relation to the electrical rotor position.

If the rotor pole is aligned with the stator tooth (0° electrical position angle or pole wheel angle) of the energized stator coil, essentially no significant motor torque can be achieved since no tangential force or transverse force can act on the rotor. On the other hand, the motor torque increases with constant control up to an offset angle of 90° electrical rotor position angle between the rotor pole and the energized stator tooth. If the offset angle is greater than 90°, the motor torque that can be achieved decreases again.

According to the invention, it is now provided to control the rotor of the electrical machine to a holding position that corresponds to an offset angle that is approximately 90°. There, the conversion between the holding current used and the motor torque that can be achieved is particularly high. Ideally, the control point can be set before an offset angle of 90° with a safety distance that is favorable in terms of control technology. In particular, the offset can be between 60" and 85°, preferably between 70" and 80°. In principle, offset angles of more than 90° are also possible Lich, which, however, can only be maintained with a lot of effort.

When holding the rotor at a holding position, the existing commutation pattern can be maintained. Due to the rotor position dependency, the measurement potential difference can be used directly as a controlled variable for the stopping position control of the electrical machine. The position control can be implemented using a PI, PD or PID controller.

Furthermore, in order to stop the electrical machine, the stator winding can be energized, in particular in unregulated or controlled operation, in such a way that the rotor is stopped in a stopping range, in particular against the action of a load torque, with the stopping range being in a range of electrical position angles in which the position angle has a predetermined safety distance from the maximum of the measurement potential difference, in particular in a range of position angles between 0° and +/-40°, in particular between 0° and +/-20°, with 0° electrical position angle of an alignment of a rotor pole to an energized one corresponds to the stator tooth.

According to one embodiment, the regulation can provide that when the rotor is stopped, it is first ensured that the rotor is in an angular range of an offset angle to a corresponding stator magnetic field, which is definitely between 0 and -45° or 0 and +45°. This can be done by impressing a high drive voltage between the two phase connections to be energized, which is done essentially independently of the current rotor position.

In particular, after stopping in the stop area, a control voltage as the difference between the applied control potentials can be reduced in absolute value continuously or step by step, so that the rotor moves in the direction of an acting load torque, depending on a course of the at least one measurement potential with regard to the electrical position angle in the course of the rotor movement, exceeding a maximum or minimum of the measurement potential is determined, after determining that the maximum or minimum of the measurement potential is exceeded, the holding position control is started.

As a result, a measurement potential difference can be continuously recorded and observed through the measurements of the measurement potentials described above by applying the measurement pulses. The control voltage is now gradually reduced, in particular by reducing the pulse duty factor for generating the control potentials. By observing the measurement potential difference, it is recognized that the rotor position is moving away from the electrical rotor position of 0° due to the load torque acting. In particular, it can be determined when the measurement potential difference exceeds a maximum or minimum. After the relevant maximum/minimum has been exceeded, the stopping position control can be activated. The measurement potential difference is used as a controlled variable, which is regulated to a value that roughly corresponds to an electrical offset angle between the stator magnetic field and the rotor magnetic field of between 60° and 85°, in particular between 70° and 80°. In this way, the most efficient possible holding of the runner at a holding position is ensured, with an electrical runner position of +/−90° being exceeded by the safety area thereby ensured, which is intended to be avoided. Holding position control is also possible with offset angles of more than 90°, but this is usually expensive to implement because the holding torque then decreases as the offset angle increases. If the positional misalignment exceeds the misalignment angle of 90°, an increased holding current would have to be applied, at least for a short time.

According to one embodiment, the stopping position control can be ended when it is determined, depending on the measurement potential, that the electrical position angle exceeds a predetermined limit angle in terms of amount or the previously exceeded maximum/minimum of the measured potential is exceeded again with the electrical position angle decreasing in terms of amount.

Provision can be made for the existing commutation pattern to be retained when the holding position control is active. In an alternative embodiment, the commutation pattern for the active holding current control can also be different from the commutation type of normal operation of the electrical machine.

• - Erfassen mindestens eines Messpotenzials, das sich nach Anlegen eines Messimpulses an einen der Phasenanschlüsse einstellt;
• - bei einem Stillstand der elektrischen Maschine, Halten des Läufers gegen ein Lastmoment mithilfe einer Haltepositionsregelung, deren Regelgröße von dem mindestens einen Messpotenzial abhängt oder diesem entspricht und die als Stellgröße die Ansteuerpotenziale der Blockkommutierung vorgibt.

According to a further aspect, a device is provided for operating an at least three-phase electronically commutated electrical machine with a stator winding having a plurality of phase strands and a rotor, the electrical machine being operated with block commutation by applying pulse-width-modulated drive potentials to phase terminals for energizing phase strands, wherein the device is designed for:

• - Detecting at least one measurement potential that occurs after applying a measurement pulse to one of the phase connections;
• - When the electrical machine is at a standstill, the rotor is held against a load torque using a holding position control, the control variable of which depends on the at least one measurement potential or corresponds to it and which specifies the control potentials of the block commutation as the manipulated variable.

According to a further aspect, a system with an at least three-phase electronically commutated electrical machine and the above device is provided.

character list

• 1 eine schematische Darstellung eines Motorsystems mit einer dreiphasig elektronisch kommutierten elektrischen Maschine mit Sternverschaltung der Statorwicklung;
• 2 ein Signal-Zeit-Diagramm der Ansteuerpotenziale bei einem gewählten Tastverhältnis und die daraus resultierenden Messpotenziale.
• 3 ein Diagramm zur Veranschaulichung eines Verlaufs der Messpotenzialdifferenz als Regelgröße einer Haltepositionsregelung abhängig von der elektrischen Läuferlage; und
• 4 ein Flussdiagramm zur Veranschaulichung eines Betriebsverfahrens für die elektrische Maschine beim Anhalten des Läufers und beim Halten des Läufers gegen ein wirkendes Lastmoment.

Embodiments are explained in more detail below with reference to the accompanying drawings. Show it:

• 1 a schematic representation of a motor system with a three-phase electronically commutated electrical machine with a star connection of the stator winding;
• 2 a signal-time diagram of the control potentials with a selected duty cycle and the resulting measurement potentials.
• 3 a diagram to illustrate a course of the measurement potential difference as a control variable of a holding position control depending on the electrical rotor position; and
• 4 a flowchart to illustrate an operating method for the electric machine when stopping the rotor and when holding the rotor against an acting load torque.

Description of Embodiments

1 shows a schematic representation of a motor system 1 with an electric machine 2, which is designed as a brushless electric machine. The electrical machine 2 can be provided as a synchronous or asynchronous motor. In the exemplary embodiment shown, the electrical machine has a three-phase design and has a stator winding 3 which, in the exemplary embodiment shown, is connected up as a star connection from phase strands 4 each consisting of one or more stator coils. However, the method described below can be operated in the same way for electrical machines with more than three phase strands and with phase strands 4 connected in a delta/polygon circuit.

In addition to the stator coils of the stator winding 3, the electrical machine 2 has a movably mounted rotor 5, which has two rotor poles 6 in the present exemplary embodiment. The following method can also be applied to electrical machines 2 with more than two rotor poles.

The stator winding 3 can be controlled via phase connections 7 . For this purpose, the phase connections are connected to a power unit 8, which can optionally apply a control potential to phase connections 7 or can switch a phase connection to zero current. The power unit 8 can be designed in a manner known per se as a B6 circuit or H-bridge circuit.

The power unit 8 is controlled via a control unit 9 . For this purpose, the control unit 9 generates control signals for transistors of the power unit 8 according to a commutation pattern.

A block commutation method can be executed in the control unit 9, which impresses a current between two of the phase connections depending on the rotor position. The current is set by specifying a drive voltage that is set between the drive potentials at at least two of the phase connections 7 and that can be generated _{by pulse width modulation of the supply voltage V vers .} The pulse width modulation is done by cyclically switching on and off transistors of the power unit depending on a duty cycle for each of the phase connections. The duty cycle determines the drive potential effectively applied to the phase terminals.

The control signals can be generated based on block commutation, with one of the phase connections being connected to a high (pulse-width modulated) supply _{potential V vers} and another of the phase connections being connected to a low (pulse-width modulated) supply _{potential V vers} , and the other phase connections being isolated from potential (by opening the with these connected transistors). The energized phase connections are selected according to the rotor position, so that the stator magnetic field is aligned in a leading direction in the direction of the motor torque to be generated.

The motor system 1 can be designed without sensors, in which case a rotor position can be determined in a manner known per se from a measurement of a measurement potential at the phase connection that is not energized. Alternatively, a position sensor for the rotor position can be provided, which has a signaling allows, which is used for a commutation of the control of the stator winding via the power unit.

Drive potentials between a high (V _vers ) and a low supply potential (0V) are preferably generated for energizing the stator winding. When a drive voltage of 0 volts is applied to the stator coils to be energized, both drive voltages are generated with a duty cycle of 50%. If a voltage greater than 0 V is to be applied, the pulse duty factor is increased from 50% for a first of the phase connections and decreased from 50% for a second of the phase connections. For example, combinations of duty cycles of 60% can be used to generate the first drive voltage and 40% to generate the second drive voltage.

The voltage pulses generated in this way at the phase connections lead to a measurement potential for the respectively unenergized phase connection, which depends on the inductance of the stator winding or on an inductance ratio of the two phase strands 4 which are energized. Since the inductance depends on the position of the rotor, the measuring potential that occurs at the de-energized phase connection can correspond to an indication of the rotor position. For this purpose, the phase connections are connected to a measuring device 10 , which measures a measurement potential at the respectively unenergized phase connection and transmits this to the control unit 9 .

Since the measurement potential at the de-energized phase connection 7 can also depend on an induction voltage that results when the rotor 5 moves, two measurement potentials M1, M2 are measured in succession for each measurement. In 2 a signal-time diagram is shown, which shows the curves of the measurement potentials M1, M2 over time with predetermined control potentials V1, V2.

The first measurement potential M1 is determined after a first control potential V1 with a rising edge from the low to the high supply potential V _{vers has been} applied to the first of the phase connections with a low supply potential at the second of the phase connections. The second measurement potential M2 is determined after a second control potential V2 with a rising edge from the low to the high supply potential V _{vers has been} applied to the second of the phase connections with a low supply potential 0 at the first of the phase connections. The measurement potentials M1, M2 are each determined after a transient period after the edge rise or fall, ie after a period after which an overshoot/undershoot due to the edge activation by the measurement pulses has subsided. The measurement potentials M1, M2 recorded in this way, which result from the opposing measurement pulses, are determined by the current rotor position. By forming the difference between the measurement potentials M1, M2, which are preferably determined in the case of control potential edges that follow one another in immediate succession, a measurement potential difference R is obtained, which can be used as a control variable for a downstream holding position control.

In 3 a diagram is shown that shows the course of the measurement potential difference ΔV as an example as a function of the electrical rotor position θ. The electric rotor position θ results from an offset angle between the directions of the stator magnetic field and the rotor magnetic field, and it is 0° in particular when a rotor pole is exactly aligned with an energized stator tooth. It can also be seen that the progression of the measurement potential difference towards offset angles of the rotor position of -90° and +90° corresponds to an approximately half-sinusoidal progression. The course of the measurement potential difference now makes it possible to use this as a controlled variable for a stopping position control of the rotor 5 at a stopping position at which the holding current can be effectively minimized. This is given as an example with CP.

The following is based on the flow chart of 4 a method is described with which a rotor of the electrical machine can be held in a holding position against the action of a load torque with a minimized holding current.

The method can be implemented in the control unit 9 as software or hardware.

In step S1 it is first checked whether the electrical machine 2 should be stopped. If this is the case, the method continues with step S2, otherwise (alternative: no) the process jumps back to step S1.

In step S2, a stopping position at a position angle in a stopping range S around a zero position (electrical position angle=0°) is approached by controlled application of a commutation pattern. This can be done, for example, by providing high motor currents with a constant commutation pattern (constant space vector), so that a rotor position angle in a range of 0° - 40° electrical rotor position between the stator magnetic field and rotor magnetic field is assumed when a positive load torque acts on the rotor 5 of the electrical machine 2 is, or so that a rotor position angle in a range around 0° - -40° electrical rotor position between the stator magnetic field and the rotor magnetic field is assumed when a negative load torque acts on the rotor of the electrical machine 2. The method is described below with reference to only a positive load torque, it also being possible to carry out the method analogously for a negative load torque. In particular, the motor current should be selected so that the rotor 5 reliably comes to a standstill with a rotor position angle in a stopping range S, ie a position angle range of 0-20° (with a rotor movement below a predetermined speed value).

In step S3 it is checked whether the rotor 5 has stopped in a range of 0° rotor position angle essentially without any significant movement. If this is determined (alternative: yes), the method continues with step S4. Otherwise (alternative: no), the process jumps back to step S2.

In step S4, the motor current is now reduced under the control of the control unit 9. This is done by gradually reducing the drive voltage (potential difference between the drive potentials) at the two selected phase connections 7. As a rule, the rotor 5 will move away from the stop area S accordingly when an external load torque acts on the rotor 5. This changes the rotor position. If no load moment acts on the rotor 5, it remains in the last position it moved to and the control potentials are reduced to 0 volts.

To apply the control voltage, the control potentials V1, V2 are generated as effective potentials, in particular pulse-width modulated, from the supply voltage as a function of a predetermined duty cycle. The voltages applied to the phase terminals 7 by the power unit 8 have square waveforms that can be used as measurement pulses. The measurement potentials M1, M2, which can be tapped after a settling time, are used to determine the measurement potential difference R.

In step S5, the measuring potentials M1, M2 of the measuring pulses are measured while the drive voltage V1, V2 is gradually being reduced, and a check is made as to whether the measuring potential difference R has exceeded a maximum or a minimum Rmin. Exceeding the maximum can be determined by recording several consecutive measurement potential differences R. In order to reduce the influence of interference signals, the course of the measurement potential differences R can be smoothed using a suitable filter. The maximum Rmax or the minimum Rmin of the measurement potential difference R is buffered.

If it is determined that the measured potential difference R decreases again after the maximum has been exceeded (alternative: yes) due to the movement of the rotor 5 due to the reduction in the control voltage, then the holding position control can be started or continued in step S6. Otherwise (alternative: no), a jump is made back to step S4 and the control voltage continues to be reduced step by step.

The holding position control is carried out with the measured potential difference R as the controlled variable and is based on a setpoint for the measured potential difference R, which corresponds to a rotor position that is determined by a rotor position angle with a control-technically favorable safety margin in front of an offset angle of 90°. In particular, the holding position control can be carried out with the measured potential difference R as the controlled variable, which corresponds to a rotor position between 60 and 85° or -60 and -85°, preferably between 70° and 80° or -70° and -80°. Alternatively, the target value of the measurement potential difference R can be dependent on a maximum or minimum measurement potential difference Rmax, Rmin, such as e.g. B. between 5 and 35% of the maximum or minimum measurement potential difference Rmax, Rmin, in particular between 10 and 25% of the maximum or minimum measurement potential difference Rmax, Rmin. The maximum or minimum measured potential difference Rmax, Rmin can be determined, for example, by continuously detecting the measured potential differences as the runner 5 moves from the stopping position to the runner position at which the runner 5 is to be held.

The holding position control can correspond to a PD, PI or PID control and provides a control voltage as a manipulated variable, which leads to the specification of corresponding pulse duty factors for the control potentials. In particular, the holding position control can directly provide the pulse duty factor for generating the control voltages V1, V2 as a manipulated variable.

In step S7, it is checked whether the absolute value of the rotor position angle has exceeded a predetermined limit angle. If it is determined during the control that the limit angle has been exceeded (alternative: yes), the control of the holding position is terminated in step S10 and the holding of the rotor 5 at the holding position is abandoned. Otherwise (alternative: no), the method is continued with step S8 and the stopping position control is continued.

In step S8, it is also checked whether due to a change in the acting load torque, in particular a reversal of direction of the load torque, the rotor 5 moves in the direction of the stop Reich moving This can be determined by measuring the potential difference R initially to the maximum Rmax rises or falls to the minimum _Rmin, this exceeds and then drops again in the direction of attitude angles in the stopping area S.

If a direction reversal is determined (alternative: yes), then in step S9 the holding position control is deactivated again after the maximum or minimum of the measurement potential difference Rmax, Rmin has been exceeded and the rotor 5 is again in a rotor position angle range of -45° to +45° . The method then continues with step S4. Otherwise (alternative: no), the system jumps back to step S6 and the stopping position control is continued.

Reference List

1

engine system

2

electric machine

3

stator winding

4

phase strand

5

runner

6

rotor poles

7

phase connection

8th

power unit

9

control unit

10

measuring device

R

measurement potential difference

Claims (10)

Method for operating an at least three-phase electronically commutated electrical machine (2) with a stator winding (3) with a plurality of phase strands (4) and a rotor (5), the electrical machine (2) being equipped with block commutation by applying control potentials (V1 , V2) is operated on phase connections (7) for energizing phase strands (4), with the following steps: - Detecting (S5) at least one measurement potential (M1, M2) that occurs after applying a measurement pulse to one of the phase connections (7); - When the electrical machine (2) is at a standstill, holding (S6) the rotor (5) against a load torque using a holding position control, the controlled variable of which depends on the at least one measurement potential (M1, M2) or corresponds to it and which, as the manipulated variable, the control potentials ( V1, V2) specifies the block commutation. procedure after claim 1 , wherein the measurement pulse is formed as part of a pulse width modulated generation of the drive potential (V1, V2), or the pulse width modulated generation of the drive potential (V1, V2) is superimposed. procedure after claim 1 or 2 , where two measurement potentials (M1, M2), in particular temporally within a pulse width modulation cycle, are detected depending on the application of two opposing measurement pulses, with a measurement potential difference (R) being determined from the two measurement potentials (M1, M2), with the holding position control as the controlled variable the measurement potential difference (R) is obtained. procedure after claim 3 , in order to stop the electric machine (2), the stator winding (3), in particular in an unregulated or controlled operation, is energized in such a way that the rotor (5) is stopped in a stopping range (S), in particular against the action of a load torque, wherein the stop range (S) lies in a range of electrical position angles in which the position angle has a predetermined safety margin to the maximum of the measurement potential difference (R), and in particular in a range of position angles of between 0° and +/-40°, in particular between 0 ° and +/-20°, with 0° electrical position angle corresponding to an alignment of a rotor pole to an energized stator tooth. procedure after claim 4 , wherein after stopping in the stop range (S), a control voltage as the difference between the applied control potentials (V1, V2) is reduced in absolute value continuously or stepwise, so that the rotor moves in the direction of an acting load torque, whereby depending on a course of the at least a measuring potential (M1, M2) with regard to the electrical position angle in the course (5) of the rotor movement, exceeding a maximum or minimum of the measuring potential is determined, after determining that the maximum or minimum of the measuring potential is exceeded, the holding position control is started. procedure after claim 5 , wherein the stop position control is configured such that the rotor of the electrical machine (2) is controlled to a stop position which corresponds to an offset angle of between 60" and 85°, preferably between 70" and 80°. procedure after claim 5 or 6 , The holding position control is ended when it is determined as a function of the measurement potential (M1, M2) that the electrical position angle exceeds a limit angle in terms of amount or that before exceeded maximum/minimum of the measurement potential (M1, M2) is exceeded again when the electrical position angle decreases in terms of absolute value. Procedure according to one of Claims 5 until 7 , whereby the existing commutation pattern is retained when the holding position control is active. Device, in particular control unit (9) for operating an at least three-phase electronically commutated electrical machine (2) with a stator winding (3) with a plurality of phase strands (4) and a rotor (5), the electrical machine (2) having block commutation by applying is operated by drive potentials generated in a pulse-width modulated manner at phase connections (7) for energizing phase strands (4), the device being designed for: - Detecting at least one measurement potential that occurs after applying a measurement pulse to one of the phase connections (7); - When the electrical machine (2) is at a standstill, the rotor (5) is held against a load torque using a holding position control, the control variable of which depends on the at least one measurement potential or corresponds to it and which specifies the control potentials of the block commutation as the manipulated variable. System with an at least three-phase electronically commutated electric machine (2) with a stator winding (3) with several phase strands (4) and a rotor (5) and a device claim 9 .

Images