Method for controlling a brushless DC motor with at least three phases

Abstract

The invention relates to a method for controlling a brushless DC motor (2) with at least three phases, preferably without a rotor position sensor, comprising: • (100) Powering on two phases (U, V, W), where the voltage of the powering is set by means of PWM control; • (200) Determine the induced voltage ΔU of an unenergized phase (U, V, W); • (400) Determining the voltage difference between the induced voltage ΔU and a reference voltage ΔU ref , where the reference voltage ΔU ref is not equal to zero • (500) Adjusting the PWM frequency if the voltage difference is not zero, wherein ◯ the PWM frequency is increased if Voltage difference at positive zero crossing of the induced voltage ΔU < 0; or Voltage difference at negative zero crossing of the induced voltage ΔU > 0; and ◯ the PWM frequency is reduced if Voltage difference at a positive zero crossing of the induced voltage ΔU > 0; or Voltage difference at the negative zero crossing of the induced voltage ΔU < 0.

Description

[0001] The invention relates to a method for controlling a brushless DC motor with at least three phases according to claim 1, a control device for controlling a brushless DC motor with at least three phases according to claim 7, a motor with a brushless DC motor with at least three phases and such a control device according to claim 13, and a household appliance with at least one such brushless DC motor with at least three phases according to claim 15.

[0002] Electric drives include synchronous motors, which are either single-phase or three-phase synchronous machines. The rotor has a constant magnetization, which, during operation, is driven synchronously by a rotating magnetic field in the stator. This results in the rotor's rotational movement within the stator. In other words, the stator's three-phase winding, typically consisting of three strands, is controlled by a suitable circuit such that a moving magnetic field is generated by the stator coils. This field pulls the rotor along, causing it to rotate. The rotational movement of the rotor in a running synchronous motor is synchronous with the alternating voltage of the stator windings. The rotor's rotational speed is related to the frequency of the alternating voltage via the number of pole pairs in the stator windings.

[0003] The rotor's magnetic field can be generated during operation using external electromagnetic excitation. This requires at least one field coil on the rotor side, which must be electrically contacted via slip rings during rotation to transfer current from the stator to the rotating rotor. This increases complexity. Furthermore, the electrical contact, usually achieved via brushes, can wear down over time. Additionally, sparking can occur at the brush ends.

[0004] On the other hand, the rotor's magnetic field can be generated by permanent magnets, a process known as self-excitation. In this case, electrical contact between the stator and rotor is unnecessary. This avoids the aforementioned disadvantages and reduces the overall weight, size, and cost of such permanent magnet synchronous motors. These permanent magnet synchronous motors with switching electronics are therefore commonly referred to as brushless motors (BLDC motors). The stator coils can be controlled via a four-quadrant controller. The electronics for controlling the bridge are a variable frequency drive. Brushless motors are typically three-phase, as this offers the advantage of a defined rotor position with high dynamics, high torque, and high efficiency.

[0005] A characteristic feature of brushless motors is their commutation, which, for example, in a three-phase stator, consists of six blocks or sectors per rotational field cycle, i.e., per motor revolution, each of which differs from the switching state of the bridge circuit. Only two push-pull stages of the bridge are active at any given time, while the third push-pull stage is in the "floating" state. The voltage at this bridge point is defined by the circuit network according to the star equivalent circuit. The bridge control ensures that the motor phase that is currently changing polarity—given a trapezoidal reverse voltage—is always in the "floating" state. Thus, in a BLDC motor, unlike in permanent magnet synchronous machines (PMSM) or PMAC motors, only two of the three phases are energized simultaneously; the third phase is de-energized.

[0006] Since the bridge control switches automatically, the stator field is always in the block with the optimal magnetic flux change, i.e., with the maximum generator voltage. The brushless motor thus spins up until its generator voltage matches the supply voltage. At that point, the maximum speed, or operating speed, is reached and maintained constant. The rotor speed can therefore be controlled or changed by varying the supply voltage. The selection of the phases to be energized depends on the rotor position. The transition from one sector to the next is also called "electronic commutation," hence the terms "brushless" or "electronically commutated."

[0007] The information about the rotor position required for the commutation of a BLDC motor is usually measured using a position sensor such as Hall sensors, resolvers or incremental encoders, which, however, leads to corresponding additional effort and costs as well as requiring installation space.

[0008] Alternatively, sensorless commutation can be implemented, eliminating the need for position sensors by indirectly determining the current rotor position through the measurement of electrical parameters at the coils. For example, the electromotive force (EMF) generated in the stator coils can be detected and evaluated by the electronic control circuit, although this is only possible above a certain minimum rotational speed.

[0009] Several solutions are known in the art that allow this commutation without a position sensor. The most widespread approach to sensorless control is based on determining the rotor position and commutation times by measuring the induced voltage in the inactive phase. The voltage potentials of the motor phase and the motor star point are compared using voltage dividers, and the zero crossing is taken as the reference time for commutation.

[0010] In brushless motors where the neutral point is not physically accessible (e.g., in a delta connection), a neutral point can be artificially created using three resistors connected in a star configuration. The voltage potential of this artificial neutral point is then compared to the voltage potential of the inactive phase, for example, using a comparator. The zero crossing of this difference is used as the reference point for commutation.

[0011] In an optimally commutated brushless motor, the phase current and the electromotive force (EMF, i.e., the induced phase voltage) should be in phase; then the operating point with the highest motor efficiency is reached. This criterion means that the zero crossing of the EMF of a phase should occur approximately in the middle of the current-free interval (assuming a small delay between voltage and current in the phase). With conventional control methods, this is not readily achieved.

[0012] DE 10 2017 117 109 A1 describes a method for controlling a brushless motor with at least three phases, preferably without a rotor position sensor, comprising: • Powering two phases, where the powering voltage is set using PWM control; • Determining the induced voltage of an unenergized phase; • Adjusting the PWM frequency if the induced voltage is not zero, where ◯ the PWM frequency is increased if the induced voltage at a positive zero crossing is < 0; or the induced voltage at a negative zero crossing is > 0; and ◯ the PWM frequency is reduced if the induced voltage at positive zero crossing is > 0; or the induced voltage at negative zero crossing is < 0.

[0013] The induced voltage is thus used to determine the commutation state. Depending on this, the PWM frequency can be adjusted to synchronize the zero crossing of the induced voltage or electromotive force with the midpoint of the current-free interval. This keeps the phase current and EMF in phase and maximizes the power-to-RMS current ratio.

[0014] In other words, the sensorless control method described in DE 10 2017 117 109 A1 determines the commutation time, or the rotor position, by measuring a phase voltage and the neutral point voltage. The induced voltage is then calculated from the difference between these two voltages. Under normal operating conditions, the induced voltage should be zero, meaning that commutation occurs synchronously with the zero crossing of the induced voltage. The phase-locked loop (PLL) of DE 10 2017 117 109 A1 synchronizes the PWM frequency with the motor speed using the voltage difference as the induced voltage, thus reducing the voltage difference to zero at the sampling time.

[0015] A disadvantage of this method, however, is that the phase voltage and the neutral point voltage can only be measured once the current in the phase has already decayed. The control method of DE 10 2017 117 109 A1 is based on the assumption that the voltages are always measured in the middle of the current-free sector. However, at high motor power levels, it can happen that the phase current has not decayed by the time of measurement in the middle of the sector.

[0016] The control method of DE 10 2017 117 109 A1 can therefore no longer be used at relatively high currents or relatively high power levels, or the current or power must be limited. This can occur particularly with low battery voltages.

[0017] The invention addresses the problem of providing a method for controlling a brushless motor of at least two phases, as described above, which can be implemented without a rotor position sensor and avoids or at least reduces the disadvantages of known control systems. In particular, the measurement point should be able to be shifted within the current-free interval at high power levels or high phase currents in order to detect the phase voltage and the neutral point voltage when the current has decayed. This should preferably be implemented in a way that is as simple, space-saving, energy-efficient, and / or cost-effective as possible. At the very least, an alternative to known methods of this kind should be provided.

[0018] According to the invention, this problem is solved by a method with the features of claim 1, by a control device with the features of claim 7, by a brushless DC motor with at least three phases and the features of claim 13, and by a household appliance with the features of claim 15. Advantageous embodiments and further developments of the invention are described in the following dependent claims.

[0019] The invention thus relates to a method for controlling a brushless DC motor with at least three phases, preferably without a rotor position sensor, comprising: • Powering two phases, where the powering voltage is set using PWM control; • Determining the induced voltage ΔU of an unenergized phase; • Determining the voltage difference between the induced voltage ΔU and a reference voltage ΔU ref, where the reference voltage ΔU ref is not equal to zero; • Adjusting the PWM frequency if the voltage difference is not zero, where ◯ the PWM frequency is increased if the voltage difference at the positive zero crossing of the induced voltage ΔU < 0; or the voltage difference at the negative zero crossing of the induced voltage ΔU > 0 and ◯ the PWM frequency is reduced if the voltage difference at the positive zero crossing of the induced voltage ΔU > 0; or the voltage difference at the negative zero crossing of the induced voltage ΔU < 0.

[0020] According to the invention, the induced voltage ΔU is used to infer the state of commutation, i.e., as known from DE 10 2017 117 109 A1.

[0021] Depending on the situation, the PWM frequency can be adjusted to synchronize the zero crossing of the induced voltage ΔU or the electromotive force (EMF) with the midpoint of the current-free interval. This keeps the phase current and EMF in phase and maximizes the power-to-RMS current ratio.

[0022] According to the invention, it is not necessary to measure all phases in the manner described. Depending on the application, it is sufficient to measure only a single phase. It is possible to measure only one phase, a subset of the phases, or all phases.

[0023] According to the invention, the voltage difference between the induced voltage ΔU and a reference voltage ΔU is further determined. ref, which is not equal to zero, so that the basic procedure or method, as known from DE 10 2017 117 109 A1, can also be applied if the current in the corresponding phase has not yet subsided at the time of measuring or tapping the voltage values.

[0024] According to one aspect of the invention, the value of the reference voltage ΔU ref predetermined and constant. This can represent a particularly simple way of implementation, although it presupposes that the reference voltage ΔU has a predetermined and fixed value. ref is chosen appropriately.

[0025] According to another aspect of the invention, the method includes the following further step: • Determining the value of the reference voltage ΔUref based on the time shift and on motor parameters, wherein the induced voltage ΔU preferably has a sinusoidal waveform.

[0026] This allows for the use of a predetermined and constant value of the reference voltage ΔU. ref This can be omitted to avoid the corresponding disadvantages. Instead, the value of the reference voltage ΔU can be ref be variable in order to adapt to the circumstances of the operation of the DC motor, which can improve the quality of the operation of the DC motor.

[0027] According to another aspect of the invention, the time shift is determined based on the current displacement angle, the sine function of the induced voltage ΔU, and the current rotational speed. This could represent a concrete implementation possibility.

[0028] According to a further aspect of the invention, the time shift is determined based on the current shift angle on the sine function of the induced voltage ΔU and interpolated values, preferably based on a look-up table. This can represent another concrete implementation possibility.

[0029] According to another aspect of the invention, the time shift is approximated based on a linear equation. This can represent another alternative, concrete possibility of implementation.

[0030] According to another aspect of the invention, determining the induced voltage ΔU comprises: • Measuring the induced voltage ΔU relative to a reference potential, where the reference potential is preferably the potential of a star point of the motor.

[0031] According to another aspect of the invention, the induced voltage of the unenergized phase is determined essentially in the middle with respect to the currentless interval.

[0032] According to another aspect of the invention, the timing of determining the induced voltage is derived from the PWM control.

[0033] According to another aspect of the invention, the PWM frequency is an integer, preferably odd, multiple of the electrical frequency.

[0034] This design allows for very simple handling of the PWM frequency or electrical frequency. Particularly in the case of an odd multiple, the trigger point can be easily derived, since a PWM pulse always occurs midway between the current-free interval. The trigger point can then be easily derived from this pulse.

[0035] According to another aspect of the invention, the method further comprises • Increasing the integer multiple if the PWM frequency falls below a specified minimum value; and / or • Decreasing the integer multiple if the PWM frequency falls below a specified maximum value.

[0036] The PWM frequency cannot be varied arbitrarily. At excessively high frequencies, unacceptable losses occur; at excessively low frequencies, the current ripple can become too strong, leading to unpleasant noise. Therefore, according to this embodiment, a minimum / maximum frequency is specified, preferably a frequency range. If threshold values ​​are exceeded, the frequency is adjusted to ensure it remains within the range or below / above the threshold values.

[0037] The present invention also relates to a control circuit for a brushless DC motor with at least three phases and without a rotor position sensor, comprising: • a control unit configured to supply power to two phases, the supply voltage being adjusted by means of PWM control; and • a scanning device designed to determine the induced voltage ΔU of an unenergized phase; where the control circuit is set up, the voltage difference between the induced voltage ΔU and a reference voltage ΔU ref to determine, where the reference voltage ΔU ref is not equal to zero wherein the control unit is set up to adjust the PWM frequency if the voltage difference is not zero, wherein ◯ the PWM frequency is increased if the voltage difference at the positive zero crossing of the induced voltage ΔU < 0; or the voltage difference at the negative zero crossing of the induced voltage ΔU > 0 and ◯ the PWM frequency is reduced if the voltage difference at the positive zero crossing of the induced voltage ΔU > 0; or the voltage difference at the negative zero crossing of the induced voltage ΔU < 0.

[0038] The inventive method can be implemented as described above by means of such a control circuit.

[0039] According to one aspect of the invention, the value of the reference voltage ΔU ref predetermined and constant.

[0040] According to another aspect of the invention, the control circuit is further configured to determine the value of the reference voltage ΔU. ref to be determined based on the time shift and on motor parameters, wherein the induced voltage ΔU preferably has a sinusoidal shape.

[0041] According to another aspect of the invention, the time shift is determined based on the current displacement angle on the sine function of the induced voltage ΔU and on the current rotational speed.

[0042] According to another aspect of the invention, the time shift is determined based on the current shift angle on the sine function of the induced voltage ΔU and interpolated values, preferably based on a look-up table.

[0043] According to another aspect of the invention, the time shift is approximated based on a linear equation.

[0044] According to another aspect of the invention, the control circuit further comprises • a microcontroller configured to provide a trigger signal to the sampling device to perform the determination of the induced voltage □U, wherein the trigger signal is output substantially midway with respect to the currentless interval.

[0045] According to another aspect of the invention, the microcontroller is configured to derive the trigger signal from the PWM control.

[0046] According to another aspect of the invention, the control unit is further configured to adjust the ratio of PWM frequency and electrical frequency of the motor such that the PWM frequency is an integer, preferably odd, multiple of the electrical frequency.

[0047] According to another aspect of the invention, the control unit is further configured to: • to increase by an integer multiple if the PWM frequency falls below a specified minimum value; and / or • to decrease by the integer multiple if the PWM frequency falls below a specified maximum value.

[0048] The present invention further relates to a brushless DC motor with at least three phases and preferably without a rotor position sensor, comprising a control circuit as described above.

[0049] According to one aspect of the invention, the brushless DC motor further comprises a star point for measuring the induced voltage.

[0050] According to another aspect of the invention • the phases are connected to the star point in a star connection; or • The star point is an artificially created star point.

[0051] The present invention further relates to a household appliance comprising a motor as described above. The household appliance can be, for example, a vacuum cleaner, but other appliances are equally suitable.

[0052] An embodiment of the invention is shown purely schematically in the drawings and is described in more detail below. It shows Fig. 1 a circuit of a brushless DC motor according to the invention; Fig. 2. Time profiles of the three phases and the corresponding induced phase voltage of the brushless DC motor according to the invention; and Fig. 3 a flowchart of a method according to the invention.

[0053] Fig. Figure 1 shows an embodiment of the invention in the form of a block diagram. A brushless motor (BLDC motor) 2 is connected to a control unit 4. The control unit 4 provides the commutation of the motor 2. The voltage applied to the phases U, V, W of the motor 2 is set by means of pulse width modulation (PWM). The control unit 4 generates the PWM signal.

[0054] Motor 2 can, for example, be controlled by a typical (not shown) circuit with a total of six power switches (two per phase), where the control unit 4 then controls the half-bridges in a corresponding pattern, i.e., opens and closes the high-side and low-side transistors of each phase U, V, W accordingly. Here and in the following, we assume a control configuration in which one phase U, V, W is energized at a time, while two phases U, V, W remain unenergized.

[0055] An ADC sampling unit, for example an analog-to-digital converter, is connected on one side to the neutral point 8 and on the other side to each of the three phases U, V, W of the motor 2. In the embodiment shown here, the motor 2 has an artificial neutral point 8, which provides the reference potential for measuring the induced phase voltage ΔU as the difference of the phase voltages U. Phase and the neutral point voltage U Star provides.

[0056] This design can be used for motors without a physical star point (e.g., in delta connection) or in star connection but without an external star point. The artificial star point 8 is created using appropriately selected resistors. Alternatively, for motors in star connection, the physically present star point can be used to tap the reference voltage.

[0057] The ADC sampling unit is triggered by a microcontroller 6 to measure the induced phase voltage at the correct time. It is advantageous for the PWM frequency to be an odd multiple of the electrical frequency, as this allows the midpoint of the PWM pulse during the current-free interval of the unenergized phase U, V, W to be selected to simultaneously trigger the measurement.

[0058] The induced phase voltage ΔU is fed back to the control unit 4 (with respect to a reference voltage ΔURef). The measured value ΔU is used as feedback via the commutation state to form a phase-locked loop (PLL) with the control unit 4. The goal of the PLL is to synchronize the zero crossing of the electromotive force (EMF) or the induced phase voltage ΔU with the sampling point in the middle of the current-free interval. This ensures that the phase current and EMF are kept in phase. The control unit 4 can be implemented entirely in hardware or at least partially in software. The PLL serves to influence the phase and, consequently, the frequency of a variable oscillator via a closed control loop in such a way that the phase deviation between an external reference signal and the oscillator, or a signal derived from it, is kept as constant as possible.

[0059] Fig. Figure 2 shows the time courses of the three phases U, V, W and the corresponding induced phase voltage ΔU.

[0060] In the Fig. Figure 2 shows the six sectors of motor 2 in BLDC operation, the control of the three phases, and the induced phase voltage ΔU in phase U. The measurement time t Mess The zero crossing of the induced phase voltage ΔU could also be measured in sector 6, or in the other two phases V and W in their respective sectors.

[0061] To determine the induced phase voltage ΔU, the phase voltage U is measured according to the invention. phase and the neutral point voltage U starSimultaneously measured in the middle of sector 3, as known from DE 10 2017 117 109 A1. Phase V continues to be clocked. However, in the previously used BLDC control system of DE 10 2017 117 109 A1, no block (1 pulse per sector) is applied to motor 2, but rather an odd number of blocks N (N ≥ 3). The phase voltage U is then measured in the middle of the central block. phase and the neutral point voltage U star measured. The induced phase voltage ΔU is then calculated as the difference between the phase voltage U and the phase voltage U. phase and the neutral point voltage U star Therefore, if N = 3, the measurement is taken in the second pulse, if N = 5, the measurement is taken in the third pulse, and so on.

[0062] For measuring the phase voltage U PhaseThe phase current must have decayed during the measurement phase (here phase U); otherwise, the voltages cannot be measured. At high power levels, however, it can happen that the phase current has not yet decayed, which is why, according to the invention, the measurement time t Mess The sampling time is shifted. According to the invention, the measurement is then taken in the last pulse; that is, with N = 3, the measurement is taken in the middle of the third pulse, and with N = 5, in the middle of the fifth pulse.

[0063] To prevent the commutation time from being shifted, the invention further regulates the induced phase voltage ΔU to a non-zero value. The setpoint for the PLL voltage, i.e., the reference voltage ΔU, is... ref , will therefore be adjusted accordingly. Various methods are conceivable for determining the target value: Regulation to a constant value other than zero Determining a value based on the time shift and the motor parameters. The induced voltage ideally has a sinusoidal waveform. The target voltage can be determined using the time shift as follows: The value on the sine function is specifically determined from the current rotational speed and the displacement angle. ΔUref=n KBemf sin(0°+Δφ) with K Bemf = Motor constant Δφ = displacement angle n = rotational speed

[0064] A look-up table containing the sine values ​​is stored, and interpolation is performed between the values ​​using the displacement angle as the input variable. ΔUref=n KBemf LUT(Δφ) with LUT = Look-Up Table

[0065] Since the sampling time is shifted by the zero crossing, the target voltage can be approximated using a linear equation. ΔUref=n KBemf mΔφ with m = constant slope

[0066] Fig. Figure 3 shows a flowchart of a method according to the invention, which is controlled by a control circuit according to the invention on the DC motor 2 according to the invention. Fig. 1 is applied.

[0067] Thus, two phases U, V, W are energized, with the energizing voltage being set via PWM control. The induced voltage ΔU of an unenergized phase U, V, W is determined. The value of the reference voltage ΔU is then determined. ref based on the time shift and motor parameters, where the induced voltage ΔU has a sinusoidal waveform. The value of the reference voltage ΔU ref It can be predetermined and constant or determined within the framework of the procedure, as described above.

[0068] In any case, the voltage difference between the induced voltage ΔU and a reference voltage ΔU is then determined. ref , where the reference voltage ΔU ref The PWM frequency is then adjusted if the voltage difference is not zero, whereby the PWM frequency is increased if the voltage difference at a positive zero crossing of the induced voltage ΔU < 0, or the voltage difference at a negative zero crossing of the induced voltage ΔU > 0, and the PWM frequency is decreased if the voltage difference at a positive zero crossing of the induced voltage ΔU > 0 or the voltage difference at a negative zero crossing of the induced voltage ΔU < 0. Reference numeral list (part of the description) U Phase U Phase V Phase W U Phase Phase voltage U star neutral point voltage ΔU induced voltage; induced phase voltage; voltage difference DU ref Reference voltage or target voltage of the induced phase voltage ΔU U MOTOR Motor voltage T PWM Pulse width modulation clock t time t Mess Measurement time; sampling time ADC scanning unit PLL phase-locked loop; phase-locked loop 2 three-phase brushless DC motors; BLDC motor 4 Control unit; computer; regulated frequency converter 6 Microcontroller 8 Star point 100 Powering two phases U, V, W 200 Determining the induced voltage ΔU of an unenergized phase U, V, W 300 Determining the value of the reference voltage ΔU ref 400 Determining the voltage difference between the induced voltage ΔU and a reference voltage ΔU ref 500 Adjusting the PWM frequency if the voltage difference is not zero QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2017 117 109 A1 [0012, 0014, 0015, 0016, 0020, 0023, 0061]

Claims

[1] Method for controlling a brushless DC motor (2) with at least three phases, preferably without a rotor position sensor, comprising: • (100) Powering on two phases (U, V, W), where the voltage of the powering is set by means of PWM control; • (200) Determine the induced voltage ΔU of an unenergized phase (U, V, W); • (400) Determining the voltage difference between the induced voltage ΔU and a reference voltage ΔU ref , where the reference voltage ΔU ref is not equal to zero • (500) Adjusting the PWM frequency if the voltage difference is not zero, wherein o the PWM frequency is increased if the voltage difference at the positive zero crossing of the induced voltage ΔU < 0; or the voltage difference at the negative zero crossing of the induced voltage ΔU > 0; and ◯ the PWM frequency is reduced if the voltage difference at the positive zero crossing of the induced voltage ΔU > 0; or the voltage difference at the negative zero crossing of the induced voltage ΔU < 0. [2] Method according to claim 1, wherein the value of the reference voltage ΔU ref is predetermined and constant. [3] Method according to claim 1, further comprising: • (300) Determining the value of the reference voltage ΔU ref based on the time shift and on motor parameters, wherein the induced voltage ΔU preferably has a sinusoidal shape. [4] Method according to claim 3, wherein the time shift is determined based on the current displacement angle on the sine function of the induced voltage ΔU and on the current rotational speed. [5] Method according to claim 3, wherein the time shift is determined based on the current shift angle on the sine function of the induced voltage ΔU and interpolated values, preferably based on a look-up table. [6] Method according to claim 3, wherein the time shift is approximated based on a linear equation. [7] Control circuit for a brushless DC motor (2) with at least three phases and without a rotor position sensor, comprising: • a control unit (4) configured to supply current to two phases (U, V, W), wherein the supply voltage is adjusted by means of PWM control; and • an ADC (automatic voltage converter) configured to determine the induced voltage ΔU of an unenergized phase (U, V, W); wherein the drive circuit is configured to measure the voltage difference between the induced voltage ΔU and a reference voltage ΔU refto determine, where the reference voltage ΔU ref is not equal to zero, wherein the control unit (4) is configured to adjust the PWM frequency if the voltage difference is not equal to zero, wherein o the PWM frequency is increased if the voltage difference at the positive zero crossing of the induced voltage ΔU < 0; or the voltage difference at the negative zero crossing of the induced voltage ΔU > 0; and ◯ the PWM frequency is reduced if the voltage difference at the positive zero crossing of the induced voltage ΔU > 0; or the voltage difference at the negative zero crossing of the induced voltage ΔU < 0. [8] Control circuit according to claim 7, wherein the value of the reference voltage ΔU ref is predetermined and constant. [9] Control circuit according to claim 7, which is further configured to determine the value of the reference voltage ΔU refto be determined based on the time shift and on motor parameters, wherein the induced voltage ΔU preferably has a sinusoidal shape. [10] Control circuit according to claim 9, wherein the time shift is determined based on the current displacement angle on the sine function of the induced voltage ΔU and on the current rotational speed. [11] Control circuit according to claim 9, wherein the time shift is determined based on the current shift angle on the sine function of the induced voltage ΔU and interpolated values, preferably based on a look-up table. [12] Control circuit according to claim 9, wherein the time shift is approximated based on a linear equation. [13] Brushless DC motor (2) with at least three phases (U, V, W) and preferably without rotor position sensor, comprising a control circuit according to one of claims 7 to 12. [14] Brushless DC motor (2) according to claim 13, further comprising a star point (8) for measuring the induced voltage ΔU. preferably • the phases (U, V, W) are connected in a star connection to the star point (8); or • the star point (8) is an artificially created star point (8). [15] Household appliance comprising a DC motor (2) according to one of claims 13 or 14.

Images