Sensorless block commutation

Abstract

The invention relates to a method for sensorless block commutation of a brushless DC motor, in which method a current sensor for detecting a current flowing through the DC motor, a detection circuit for detecting the zero crossings of the counter-electromotive force, and a control unit are used, the control unit being designed to provide, based on the detected current and / or the zero crossings of the counter-electromotive force, block drive signals and position detection signals for a motor inverter associated with the DC motor, the block drive signals being used to generate a drive force in order to rotationally drive a rotor of the DC motor, and the position detection signals being used to detect a rotor position of the rotor, and the DC motor passing through a plurality of rotor position sectors in a start-up phase before reaching a normal operating phase, and the control unit outputting exclusively position detection signals to the motor inverter in the start-up phase.

Description

[0001] Sensorless block commutation

[0002] The present invention relates to a method for sensorless block commutation of a brushless DC motor. Within the scope of the method, a current sensor for detecting a current flowing through the DC motor, a detection circuit for detecting the zero crossings of the counter-electromotive force, and a control unit are used. The control unit is suitable for providing block control signals and position detection signals for a motor inverter assigned to the DC motor based on the detected current and / or the zero crossings of the counter-electromotive force. The block control signals typically serve to generate a drive force to rotate a rotor of the DC motor. The position detection signals typically serve to detect a rotor position of the rotor. Within the scope of the method, the DC motor undergoes a start-up phase before reaching a normal operating phase.It has been shown in practice that during the start-up phase, it is not possible to reliably determine a zero crossing of the counter electromotive force (BEMF zero crossing detection) - as a basis for sensorless block commutation.

[0003] Devices and methods for sensorless block commutation are generally known from the state of the art.

[0004] US 9,819,290 B2 describes a power tool with a motor and a control unit configured to output detection signals and control signals for an inverter. The control unit determines the block control signals based on position detection current components of a motor current for mode control during a start-up phase of the motor.

[0005] US 1 1 ,171 ,586 B2 discloses a power tool with a control unit designed to detect an incorrect rotor rotation direction by applying a first plurality of voltage pulses to a current motor sector and a second plurality of voltage pulses to a previous sector.

[0006] The object of the present invention is to provide a method for sensorless block commutation of a DC motor that creates the basis for more precise tracking of the DC motor's rotor position. This object is achieved by the control unit exclusively outputting position detection signals during the start-up phase. In other words, during the start-up phase, the DC motor is rotated solely by position detection currents triggered by the position detection signals.

[0007] The invention incorporates the finding that, in previously known commutation methods, a current signal detected by the current sensor during the start-up phase contains both control current components and position detection current components. While this allows for a high starting torque, it reduces the accuracy of rotor position tracking.

[0008] The invention further includes the finding that stator inductance changes can be exploited in sensorless motor control to detect and track the rotor position below the limit at which tracking based on BEMF-based methods is possible. Two fundamental physical properties of the motor can be exploited in this process: stator saturation effects and / or salience effects. A change in the current signal detected by the current sensor influenced by these properties preferably forms the basis of the method according to the invention.

[0009] In a particularly preferred embodiment, the position detection signals have a plurality of main pulses and a plurality of secondary pulses. It has proven advantageous if main pulses and secondary pulses alternate. A main pulse and a secondary pulse can have the same pulse duration. Alternatively, a main pulse and a secondary pulse can have different pulse durations. It has proven advantageous if the same pause times are provided between the transition from main pulse to secondary pulse or from secondary pulse to main pulse. Alternatively, a different pause time can be provided between the transition from main pulse to secondary pulse or from secondary pulse to main pulse. An alternating sequence of main pulses and secondary pulses is defined as a pulse sequence.

[0010] In a particularly preferred embodiment, a control state assigned to the current rotor position sector is used for the main pulse, and a control state assigned to a next rotor position sector immediately following the current rotor position sector in the direction of rotor rotation is used for the secondary pulse. In a further particularly preferred embodiment, a valid change of the rotor position sector is detected when the magnitude of a main pulse current response to the main pulse detected by the current sensor is greater than the magnitude of a secondary pulse current response to the secondary pulse detected by the current sensor. It has proven advantageous if the main pulse current response to the main pulse detected by the current sensor exhibits a commutation current hysteresis when compared with the secondary pulse current response to the secondary pulse detected by the current sensor.A valid change in the rotor position sector is only detected when the magnitude of the main pulse current response is greater than the secondary pulse current response by the commutation current hysteresis. This commutation current hysteresis can be adjustable. In a particularly preferred embodiment of the method, after detection of a valid change in the rotor position sector, a preferably adjustable pause time is waited for, during which the motor phase currents completely decay. After the pause time, the next pulse sequence is started.

[0011] It has proven advantageous that, when a valid change in the rotor position sector is detected, the control states of the main pulse and the secondary pulse are incremented or decremented, i.e., switched to the next rotor position sector. It has proven advantageous if switching only occurs if a minimum number of main pulses and / or secondary pulses have been impressed per pulse sequence. Alternatively or additionally, it can be provided that switching only occurs if a maximum number of pulse impressions of main pulses and / or secondary pulses per pulse sequence has not been exceeded. In a further particularly preferred embodiment, the current rotor speed itself is determined over the period of time during which a pulse sequence is impressed. Each pulse sequence can be valid for one rotor position sector, preferably 60°.From these two pieces of information, if the number of pole pairs of the motor is known, the mechanical speed can be clearly determined.

[0012] In a further particularly preferred embodiment, it is provided that when a valid change of the rotor position sector is detected, a direction detection pulse is output once as a position detection signal, wherein a control state is used for the direction detection pulse which is assigned to a rotor position sector immediately preceding the current rotor position sector in the direction of rotor rotation.

[0013] In a particularly preferred embodiment, at the beginning of the start-up phase, with the rotor stationary, an initial rotor position detection takes place which does not generate a resulting rotor torque.

[0014] It has proven advantageous if the normal operating phase begins as soon as the rotor speed reaches a value within the range of 1000 rpm to 3000 rpm. In a particularly preferred embodiment, commutation of the brushless DC motor during the normal operating phase occurs exclusively by means of block commutation. In another particularly preferred embodiment, commutation of the DC motor during the normal operating phase occurs based on a determination of the zero crossings of the counter electromotive force.

[0015] The object is also achieved by an electric handheld power tool with a brushless DC motor, a current sensor for detecting a current flowing through the DC motor, a detection circuit for detecting the zero crossings of the counter-electromotive force, and a control unit. The control unit is designed to provide, based on the detected current and / or the zero crossings of the counter-electromotive force, block control signals and position detection signals for a motor inverter assigned to the DC motor. The block control signals serve to generate a drive force to rotate a rotor of the DC motor, and the position detection signals serve to detect a rotor position of the rotor. The DC motor is further designed to pass through several rotor position sectors in a start-up phase before reaching a normal operating phase.According to the invention, the control unit is designed to output exclusively position detection signals to the motor inverter during the start-up phase.

[0016] The electric handheld power tool can be further developed in a corresponding manner by the features described with reference to the method. The method according to the invention can also be implemented in other devices equipped with a brushless DC motor, such as a vacuum cleaner or chipping hammer.

[0017] In a particularly preferred embodiment, the brushless DC motor is 3-phase and has six sectors of 60° each in the electrical reference system. Accordingly, the motor inverter is designed to commutate a 3-phase DC motor. This motor inverter thus has three half-bridges, each with two power semiconductors per half-bridge. It has proven advantageous if the brushless DC motor is connected in a delta configuration. Alternatively, the brushless DC motor can be connected in a star configuration. Further advantages will emerge from the following description of the figures. Various exemplary embodiments of the present invention are illustrated in the figures. The figures, the description, and the claims contain numerous features in combination. Those skilled in the art will expediently also consider the features individually and combine them into useful further combinations.

[0018] In the figures, identical and similar components are numbered with the same reference numerals. They show:

[0019] Fig. 1 shows a first embodiment of a hand-held power tool according to the invention;

[0020] Fig. 2 shows a first embodiment of a method according to the invention; Fig. 3 shows a second embodiment of a method according to the invention;

[0021] Fig. 4 shows an initial rotor position detection in the process of Fig. 2; and

[0022] Fig. 5 shows an alternative method for initial rotor position detection.

[0023] Examples of implementation:

[0024] Fig. 1 shows—in highly schematic form—an electric handheld power tool 100. The handheld power tool 100 is equipped with a brushless DC motor 90. The brushless DC motor 90 is a three-phase motor and has a rotor 95 with six rotor position sectors I, II, III, IV, V, VI, each with a pitch of 60°. Associated with the DC motor 90 is a current sensor 80 for detecting a current flowing through the DC motor 90. Additionally, a circuit for zero-crossing detection of the counter-electromotive force 70 is provided.

[0025] A motor inverter 20 is connected upstream of the DC motor 90, which in this case has six FETs (field effect transistors), with two FETs being provided per phase II, V, W. The electric hand-held power tool 100 further comprises a control unit 10, which is designed to provide block control signals and position detection signals for the motor inverter 20 based on the current detected by the current sensor 80 and / or the zero crossings of the counter electromotive force. The block control signals serve (in a manner known from the prior art) to generate a driving force in order to drive the rotor 95 in rotation. The position detection signals serve to detect a rotor position of the rotor 95. According to the invention, the control unit 10 is designed, in a start-up phase, ieFor example, up to a speed of the DC motor of 2000 ll / min, only position detection signals are output to the motor inverter 20, which is explained in more detail with reference to the following figures.

[0026] Fig. 2 shows a diagram of the time course of the phase currents A, B, and C (corresponding to phases II, V, and W in Fig. 1) detected by three current sensors. The diagram is divided into an initial position detection / rest phase IPD / RPH, a start-up phase APH, and a normal operation phase NPH. During the initial position detection / rest phase IPD / RPH, the rotor is stationary. The initial rotor position detection will be explained in more detail later with reference to Figs. 4 and 5.

[0027] In the start-up phase APH, which immediately follows the initial position detection and extends, for example, to approximately 30 milliseconds, it is not reliably possible to determine a zero crossing of the back electromotive force (BEMF zero-crossing detection) as the basis for the sensorless block commutation of the DC motor 90. This is where the invention comes into play, which provides that the control unit outputs only position detection signals to the motor inverter during the start-up phase APH. Within the scope of the method, two voltage pulses, namely a main pulse and a secondary pulse, are alternately impressed as position detection signals on the motor inverter 20, and the resulting current amplitudes are recorded and compared. Therefore, this method can also be referred to as the "2-Pulse Acceleration Method." The impressed main pulses and secondary pulses generate a resulting torque that causes the rotor to rotate.

[0028] Accordingly, the diagram in Fig. 2 shows the main pulse current responses l(Pi) alternating with the secondary pulse current responses l(P2). For the main pulse, a control state of the motor inverter assigned to the current rotor position sector is used. For the secondary pulse, a control state is used that is assigned to a rotor position sector immediately following the current rotor position sector in the direction of rotor rotation. The main pulse current responses l(Pi) and the secondary pulse current responses l(P2) change with the rotor position, which is due to the saturation effects and salience effects mentioned above.

[0029] If the main pulse current response l(Pi) of the main pulse has a greater amplitude than the secondary pulse current response l(P2) of the secondary pulse, the rotor is in the next rotor position sector. For robustness reasons, an adjustable commutation current hysteresis iHys is also used, which is applied to the secondary pulse current response l(P2). Accordingly, a valid change of the rotor position sector is detected if the condition main pulse current response l(Pi) > secondary pulse current response l(P2) +

[0030] Commutation current hysteresis IHys is met. If such a valid change in the rotor position sector is detected, the control states of the main pulse and the secondary pulse are incremented or decremented. The first pulse sequence PSi is completed. Subsequently, an adjustable pause time TPS is waited for. During this pause time TPS, the motor phase currents must completely decay, as shown in Fig. 2. Once the pause time TPS has expired, the next pulse sequence, in this case the second pulse sequence PS2, begins. In the start-up phase (APH) shown in Fig. 2, there are eight pulse sequences as an example. At the end of the eighth pulse sequence PS8, there is a switching point in time (UP) at which the normal operating phase (NPH) of the DC motor begins. From this switching point in time (UP), commutation of the brushless DC motor takes place exclusively by means of block commutation.The commutation of the DC motor in the normal operating phase is based on a determination of the zero crossings of the counter electromotive force.

[0031] Fig. 3 shows the so-called "3-Pulse Acceleration Method," in which, in addition to the method described with reference to Fig. 2, a direction detection pulse is output to the motor inverter as a position detection signal once in each pulse sequence (PS). Each such direction detection pulse causes a direction detection pulse current response l(P3) shown in Fig. 3. For each direction detection pulse, a control state is used that is assigned to a rotor position sector immediately preceding the current rotor position sector in the direction of rotor rotation. A direction detection pulse is applied whenever a valid change in the rotor position sector is detected, as described in Fig. 2.

[0032] If the direction detection pulse current response l(P3) of the direction detection pulse has a greater amplitude than the last measured main pulse current response l(Pi) of the main pulse, the rotor rotates in the desired direction. For robustness reasons, an adjustable direction current hysteresis iHysRi is also used, which is applied to the main pulse current response l(Pi). Accordingly, a rotation of the rotor in the desired direction is detected if the condition direction detection pulse current response l(P3) > main pulse current response I(P1) + direction current hysteresis IHysRi is met. Otherwise, the rotor rotates in the undesired direction.

[0033] The methods described with reference to Figs. 2 and 3 can each be supplemented by an initial rotor position detection. This will now be described in more detail with reference to Figs. 4 and 5.

[0034] Figs. 4 and 5 each show a diagram of the temporal progression of the phase currents A, B, and C (corresponding to phases II, V, and W in Fig. 1) measured by three current sensors. The initial rotor position detection occurs during a rest phase (RPH), which precedes the start-up phase (APH) (see Figs. 2 and 3). During the rest phase (RPH), the rotor is at a standstill.

[0035] Initial rotor position detection is preferably performed by applying pairs of opposing control states to avoid generating a resulting torque on average. Pause times are provided after the pulse application so that the generated currents can completely decay before the next pulse is applied.

[0036] During initial rotor position detection, the same physical properties are preferably used that are also used during the start-up phase (APH). During initial rotor position detection, differences in the current responses of the impressed voltage pulses are used to infer the initial rotor position. However, at certain rotor positions, more precisely the detection sector boundaries, there is an angular uncertainty of 60° (one rotor position sector). At these positions, even measurement noise from the current measurement can lead to incorrect detection of the initial rotor position by one rotor position sector. In order to minimize the frequency of incorrect determination of the initial rotor position, it has proven advantageous if the detent positions of the brushless DC motor do not coincide with the detection sector boundaries of the method.This can be explained by the fact that the magnetic system of the DC motor strives for the lowest magnetic resistance (reluctance), and this is achieved at the motor's magnetic detent positions. Thus, the detent positions represent the most common rotor positions during initial rotor position detection. Furthermore, the orientations of the impressed voltage vectors are determined by the motor's wiring (star or delta). The orientations of the impressed voltage vectors differ by 30° between a star and a delta connection.

[0037] The method and / or device according to the invention can provide for the use of 3-FET block commutation or 2-FET block commutation, depending on the electrical wiring of the brushless DC motor (star or delta). These two control systems differ in that they have different control states and thus impose voltage pulses that differ by 30°. This compensates for the difference in wiring. The detection limits can thus always be separated from the magnetic detent positions. This offers the advantage that the detection algorithm described above can remain unchanged. In this regard, the invention includes the discovery of using different control states to separate the detection limits of the method from the magnetic detent positions of the motor, depending on the wiring.

[0038] Fig. 4 shows an initial rotor position detection for a brushless DC motor in a star connection, which is configured as a three-phase motor with six sectors, each with an angle of 60° in the electrical reference system. For this motor configuration, it is advantageous for the initial rotor position detection if the motor inverter has three active power semiconductors (3-FET) during a position detection pulse.

[0039] Fig. 5 shows an initial rotor position detection for a brushless DC motor in a delta connection, which is designed as a three-phase motor with six sectors, each with a 60° angle in the electrical reference system. For this motor configuration, it is advantageous for the initial rotor position detection if the motor inverter has two active power semiconductors (2-FET) during a position detection pulse.

[0040] 10 Control unit

[0041] 20 motor inverters

[0042] 70 BEMF zero-crossing detection circuit

[0043] 80 current sensor

[0044] 90 brushless DC motor (BLDC motor)

[0045] 95 Rotor

[0046] 100 Electric hand tool

[0047] I, II ... VI rotor position sectors

[0048] IPD Initial rotor position detection

[0049] APH start-up phase

[0050] I(P1) Main pulse current response l(P2) Subpulse current response l(P3) Direction detection pulse current response

[0051] IHys commutation current hysteresis

[0052] IHysRi Directional current hysteresis

[0053] NPH normal operating phase

[0054] RPH resting phase

[0055] PS1, PS2 pulse sequence

[0056] U, V, W phases of the DC motor

[0057] UP switching time

[0058] TPs break time

Claims

Claims: A method for sensorless block commutation of a brushless DC motor (90), wherein the method includes a current sensor (80) for detecting a current flowing through the DC motor (90), a detection circuit (70) for detecting the zero crossings of the counter-electromotive force, and a control unit (10), wherein the control unit (10) is suitable for providing block control signals and position detection signals for a motor inverter (20) associated with the DC motor (90) based on the detected current and / or the zero crossings of the counter-electromotive force, wherein the block control signals serve to generate a drive force to rotate a rotor (95) of the DC motor (90), and the position detection signals serve to detect a rotor position of the rotor (95).and wherein the DC motor (90) passes through several rotor position sectors in a start-up phase (APH) before reaching a normal operating phase (NPH), characterized in that the control unit (10) outputs position detection signals exclusively to the motor inverter (20) during the start-up phase (APH). Method according to claim 1, characterized in that the position detection signals comprise a single or multiple main pulses and a single or multiple secondary pulses. Method according to claim 2, characterized in that the main pulses and secondary pulses alternate. Method according to claim 2 or 3, characterized in that a control state associated with the current rotor position sector is used for the main pulse and a control state associated with a next,The rotor position sector immediately following the current rotor position sector in the direction of rotor rotation is assigned to the rotor position sector. A method according to one of claims 2 to 4, characterized in that a valid change of the rotor position sector is detected when the magnitude of a main pulse current response detected by the current sensor (80) I(P1) on the main pulse is greater than the magnitude of a secondary pulse current response l(P2) to the secondary pulse detected by the current sensor (80).

6. Method according to claim 5, characterized in that, when a valid change of the rotor position sector is detected, the control states of the main pulse and the secondary pulse are incremented or decremented.

7. Method according to claim 5 or 6, characterized in that, when a valid change of the rotor position sector is detected, a direction detection pulse (P3) is output once as a position detection signal, wherein a control state is used for the direction detection pulse which is assigned to a rotor position sector immediately preceding the current rotor position sector in the direction of rotor rotation.

8. Method according to at least one of the preceding claims, characterized in that at the beginning of the start-up phase (APH), with the rotor (95) stationary, an initial rotor position detection (IPD) is carried out which does not generate a resulting rotor torque.

9. Method according to at least one of the preceding claims, characterized in that the normal operating phase (NPH) begins as soon as the rotor speed reaches a value in the interval of 1000 l / min to 3000 l / min.

10. Method according to at least one of the preceding claims, characterized in that the commutation of the brushless DC motor (90) in the normal operating phase (NPH) is carried out exclusively by means of block commutation.

11. Method according to at least one of the preceding claims, characterized in that the commutation of the DC motor (90) in the normal operating phase (NPH) is carried out on the basis of a detection of the zero crossings of the back electromotive force. Electric hand tool (100) comprising a brushless DC motor (90), a current sensor (80) for detecting a current flowing through the DC motor (90), a detection circuit (70) for detecting the zero crossings of the counter-electromotive force, and a control unit (10), wherein the control unit (10) is configured to provide block control signals and position detection signals for a motor inverter (20) associated with the DC motor (90) based on the detected current and / or the zero crossings of the counter-electromotive force, wherein the block control signals serve to generate a drive force to rotate a rotor (95) of the DC motor (90), and the position detection signals serve to detect a rotor position of the rotor (95), and wherein the DC motor (90) is configured to perform several rotor position sectors (I, II ...) before reaching a normal operating phase (NPH).VI) to be traversed in a start-up phase (APH), characterized in that the control unit (10) is designed to output position detection signals exclusively to the motor inverter (20) during the start-up phase (APH).

Images