Control device for a polyphase motor and method for driving a polyphase motor

Abstract

An actuating apparatus for a polyphase motor includes five phase terminals for connecting of in each case one phase of the polyphase motor, a high terminal for applying a supply voltage (U.sub.B), a low terminal for applying a reference potential of the supply voltage (U.sub.B), a control device, wherein the control device is adapted, in four of the five phase terminals, to impress a pulse-width-modulated voltage pattern by connecting the phase terminals to the high terminal or the low terminal so that in the first phase terminal, an evaluation signal which is dependent on the angle of rotation of the polyphase motor is produced, and wherein the control device is adapted to determine the angle of rotation and/or a commutation condition of the polyphase motor from the evaluation signal.

Claims

1. An actuating apparatus for a polyphase motor, comprising: five phase terminals for connecting in each case a phase of the polyphase motor; a high terminal for applying a supply voltage (U.sub.B); a low terminal for applying a reference potential of the supply voltage (U.sub.B); a control device; wherein the control device is adapted, in one operational state, to impress a pulse-width-modulated voltage pattern in four of the five phase terminals by connecting the four phase terminals to the high terminal or the low terminal through at least two consecutive switching states, while switching a remaining fifth of the five phase terminals to passive for the at least two consecutive switching states, thereby producing an evaluation signal which is dependent on the angle of rotation of the polyphase motor in the remaining fifth of the five phase terminals; and wherein the control device is adapted to determine the angle of rotation and/or a commutation condition of the polyphase motor from the evaluation signal.

2. The actuating apparatus according to claim 1, further comprising a plurality of bridge branches, wherein each bridge branch is connected to precisely one of the five phase terminals in order to impress the pulse-width-modulated voltage pattern in the four of the five phase terminals.

3. The actuating apparatus according to claim 2, wherein each of the bridge branches comprises a series connection of a high switch and a low switch; wherein the high switch of each of the bridge branches is connected to the high-terminal; wherein the low switch of each of the bridge branches is connected to the low terminal-; wherein each of the five phase terminals is connected between the high switch and the low switch of the respective bridge branches.

4. The actuating apparatus according to claim 1, wherein the control device is adapted to impress the pulse-width-modulated voltage pattern in the four of the five phase terminals so as to connect half of the four phase terminals to the high terminal, and to connect another half of the four phase terminals to the low terminal.

5. The actuating apparatus according to claim 1, wherein the pulse-width-modulated voltage pattern is periodically passed through with a predetermined step duration during a predeterminable first time period (T1); wherein the pulse-width-modulated voltage pattern is adapted in such a way that connecting of the phase terminals to the high terminal or the low terminal per phase terminal is maintained substantially over two step durations of the pulse-width-modulated voltage pattern.

6. The actuating apparatus according to claim 1, wherein the evaluation signal has two sinusoidal voltage progressions having a phase shift of 90.

7. The actuating apparatus according to claim 5, wherein the evaluation signal has a first sinusoidal voltage progression from a voltage differential between two temporally interrupted step durations of the impressed pulse-width-modulated voltage pattern.

8. The actuating apparatus according to claim 7, wherein the evaluation signal has a second sinusoidal voltage progression from a voltage differential between two additional non-contiguous step durations of the impressed pulse-width-modulated voltage pattern; wherein the two additional non-contiguous step durations have step durations of the pulse-width-modulated voltage pattern which are different from the step durations on which the first sinusoidal voltage progression is based.

9. The actuating apparatus according to claim 7, wherein the control device is adapted to determine the angle of rotation of the polyphase motor from the first sinusoidal voltage progression and the second sinusoidal voltage progression.

10. The actuating apparatus according to claim 1, wherein, after a predeterminable first time period, a phase terminal other than the first phase terminal is used to determine the evaluation signal.

11. The actuating device according to claim 1, further comprising: at least two additional phase terminals for connecting in each case one additional phase of the polyphase motor.

12. A motor control system, comprising: the actuating apparatus according to claim 1; and a motor having at least five phases; wherein in each case one of the five phase terminals of the actuating apparatus is connected to one of the at least five phases.

13. A method for actuating a polyphase motor, comprising: applying a supply voltage (U.sub.B) to a high terminal of an actuating apparatus according to claim 1; applying a reference potential of the supply voltage (U.sub.B) to a low terminal of the actuating apparatus; impressing a pulse-width-modulated voltage pattern in four of the five phase terminals of the actuating apparatus by connecting the phase terminals with the high terminal or the low terminal for at least two consecutive switching states while switching a fifth of the five phase terminals to passive for the at least two consecutive switching states; detecting in the fifth phase terminal an evaluation signal which is dependent on the angle of rotation of the polyphase motor; and determining an angle of rotation and/or a commutation condition of the polyphase motor from the evaluation signal.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) In the following, further exemplary embodiments of the present invention will be described with reference to the drawings.

(2) FIG. 1 is a schematic block diagram of an actuating apparatus for a motor according to an exemplary embodiment of the present invention.

(3) FIG. 2 shows a rotor coordinate system in relation to a stator coordinate system of an electric motor according to an exemplary embodiment of the present invention.

(4) FIG. 3 shows the progression of two voltage differentials in a passive phase terminal during rotation of the rotor according to an exemplary embodiment of the present invention.

(5) FIG. 4 shows the progression of an absolute voltage in a passive phase with corresponding switching states according to an exemplary embodiment of the present invention.

(6) FIG. 5 shows a detail from FIG. 4 according to an exemplary embodiment of the present invention.

(7) FIG. 6 shows a detail from FIG. 4 in which the rotor is rotated 90 further relative to FIG. 5 according to an exemplary embodiment of the present invention.

(8) FIG. 7 shows the sequence of the ten-step commutation according to an exemplary embodiment of the present invention.

(9) FIG. 8 is a flow chart for a method for actuating a motor according to an exemplary embodiment of the present invention.

DETAILED DESCRIPTION OF EMBODIMENTS

(10) The drawings in the figures are schematic and not to scale. In the following description of FIG. 1 to FIG. 8, the same reference signs are used for the same or corresponding elements.

(11) FIG. 1 is a schematic block diagram of an actuating apparatus 100 for a motor 140, M according to an exemplary embodiment of the present invention. The motor 140 has five phases 141, 142, 143, 144, 145 having corresponding phase windings a, b, c, d, e. The five phases 141, 142, 143, 144, 145 are interconnected at a common star point Y. The star point Y can, as indicated in FIG. 1, be brought out of the motor housing through a star line 150. However, the connection 150 is often not present. The phases 141, 142, 143, 144, 145 are connected to corresponding phase terminals 111, 112, 113, 114, 115 of the actuating apparatus 100. The star point Y can also be reached via each of the phase terminals 111, 112, 113, 114, 115.

(12) The control apparatus 100 additionally comprises a high terminal 102 and a low terminal 103 for applying a battery voltage U.sub.B (not shown in FIG. 1). The high terminal 102 is connected to a first high switch 104, a second high switch 104, a third high switch 104, a fourth high switch 104 and a fifth high switch 104. A terminal of the high switches 104, 104, 104, 104, 104 which in each case is opposite the terminal of the high switch 104, 104, 104, 104, 104 which is connected to the high terminal 102, is in each case connected to one of the phase terminals 111, 112, 113, 114, 115. Using this terminal, the series connection of the high switch and the low switch of each bridge branch 125, 126, 127, 128, 129 is produced. The first high switch 104 is consequently connected to the first phase terminal 111, the second high switch 104 is connected to the second phase terminal 112, the third high switch 104 is connected to the third phase terminal 113, the fourth high switch 104 is connected to the fourth phase terminal 114, and the fifth high switch 104 is connected to the fifth phase terminal 115. In addition, the first high switch 104 is connected to a first low switch 105, the second high switch 104 is connected to a second low switch 105, the third high switch 104 is connected to a third low switch 105, the fourth high switch 104 is connected to a fourth low switch 105, and the fifth high switch 104 is connected to a fifth low switch 105. At the respective connection points at which each high switch is connected to the corresponding low switch, the phase terminal 111, 112, 113, 114, 115 and a measuring terminal 120, 121, 122, 123, 124 are also connected. The measuring terminal 120, 121, 122, 123, 124 can be used to measure voltages which are induced in the phases and to feed the measured voltages back to the control device 109. Said reference potential 103 can be for example a ground potential, GND or a ground terminal 103. In parallel with the high and/or low switches, i.e. in parallel with the bridge branches 125, 126, 127, 128, 129 or the half bridges 125, 126, 127, 128, 129, a capacitor 106 is arranged. Said capacitor is used to smooth the PWM signals generated by the bridge branches 125, 126, 127, 128, 129.

(13) The five phase terminals 111, 112, 113, 114, 115 are connected to five sampling terminals 120, 121, 122, 123, 124, five sampling connections 120, 121, 122, 123, 124 or five measuring terminals 120, 121, 122, 123, 124. These lead to an evaluation device 107. The evaluation device 107 is connected to a control unit 109 or control device 109 via a connection line 108 or feedback line 108. The control device 109 is connected to the high switches 104, 104, 104, 104, 104 and low switches 105, 105, 105, 105, 105 via the switching terminals 110. Each of the switches has its own physical connection 110 to the control device. The control line 110 can alternatively be in the form of a bus so that each switch has a logical connection to the control device 109. The control device 109 is connected to the switches 104, 104, 104, 104, 104, 105, 105, 105, 105, 105 to actuate or drive the bridge branches. The combination of high switches 104, 104, 104, 104, 104 and low switches 105, 105, 105, 105, 105 form five bridge branches 125, 126, 127, 128, 129. Each of the bridge branches is thus connected to one of the phase terminals 111, 112, 113. The switches 104, 104, 104, 105, 105, 105 can be implemented by means of transistors or electronic switches.

(14) The control device 109 or the processor 109 is adapted in such a way that both the high switch of each of the bridge branches and the low switch of each of the bridge branches can be actuated in a predeterminable sequence, the actuation taking place in such a way that one of the phase terminals is switched to passive. A phase terminal 111 switched to passive means that said phase terminal 111 is separated from the supply terminals in each case by means of the high switch 104 and the low switch 105, that is to say is separated from the terminals of the supply voltage 102 and the reference potential 103.

(15) In the example in FIG. 1, the first phase terminal 111 which belongs to a phase coil a of the phase 141 of the motor 140 is switched to passive. The phase terminal switched to passive is shown by a dashed line in FIG. 1. By switching off the switches 104 and 105, the sampling input 121, which belongs to the phase 111, 141 switched to passive, can be used to measure a voltage induced in the phase coil a. In particular, by using phase a, switched to passive, the self-induction of the parallel connection of two active phase coils is measured between the branch point Y and the reference potential 103. This fact about the phase coil a, switched to passive, is represented in Table 1 by an O. Phase coils switched to active are denoted by L or H depending on the switching state thereof. H means that the corresponding phase coil, phase and/or the phase terminal is connected to the supply voltage terminal 102 via the corresponding high switch 104, 104, 104, 104, 104 and is thus also connected to the supply voltage U.sub.B. L means that the corresponding phase coil, phase and/or the phase terminal is connected to the reference potential terminal 103 via the corresponding low switch 105, 105, 105, 105, 105 and is thus also connected to the reference potential.

(16) TABLE-US-00001 TABLE 1 T1 S1 S2 S3 S4 a O O O O b H L L H c H H L L d L H H L e L L H H

(17) Alternately, by using the control device 109, in each case one of the phases 141, 142, 143, 144, 145 and/or one of the phase coils a, b, c, d, e can be switched to passive. The phases are switched to passive according to the angle of rotation. The pre-set angle for the commutation angle in this case means that the angles are each 360/10, i.e. 36, up to the next step. In the case of five phases, a ten-step commutation is carried out. In the case of a three-phase machine with six-step commutation, the spacing of the commutation angles is 60=360/6. The switches 104, 104, 104, 104, 105, 105, 105, 105, which belong to the active coils, are actuated periodically by using the control device 109 to generate an alternating signal, in particular a PWM (pulse-width-modulated) signal. The switch is operated in such a way that, within an active bridge branch, the high switch and the low switch are switched in opposite directions so that, within an active bridge branch, always just precisely one of the two switches is connected and produces a new connection. Only in the case where a phase terminal 111, 112, 113, 114, 115 is switched to passive are both the corresponding high switch and the corresponding low switch switched off, open and not involved.

(18) The motor 140, M is used in a ten-step commutation mode. The four active motor phase terminals 112, 113, 114, 115 and the corresponding active phase coils b, c, d, e are used to generate a voltage vector. Induced voltages result from the change in current effected by the PWM and the self-inductance of the coils involved. The induced voltages are measured when at least two active phases are connected to U.sub.B, i.e. when at least two phases are switched to the high state (H), and the two other phases are connected to GND, thus are in the low state (L).

(19) Instead of one passive phase and four active phases, it is also possible to use only two of the four active phases to determine the angle. However, then the full efficiency of the motor would not be achieved, since only part of the motor is used, namely two active and three passive phases, and since the signals in the three passive phases are substantially the same. In one example, however, minimal differences in the three passive phases can be detected by means of precise measurement technology, and the detected differences can be used for further motor characterisation and to describe the magnetic coupling.

(20) By setting the switches in the respective states, parallel connections of the connected phases are produced. A voltage divider is formed, the centre of which is the star point Y or branch point Y.

(21) When considering for example the state S1 from Table 1, the high switches 104, 104 are in the state H and the low switches 105, 105 are in the state L. Therefore, the coils b, c are coils which are switched to H, and the coils e, d are coils which are switched to L. The coils b, c and e, d are connected in parallel and form a voltage divider for the supply voltage U.sub.B. In the example, this means that in the state S1, the coils b, c are connected in parallel, and the coils d, e. By varying the states, on the premise that two switches must be actuated from state to state, four states can be passed through S1, S2, S3, S4. The states are selected in this case in such a way that the switching states of precisely two coils always change from state to state so that always half of the number of the active phase coils are connected to H, and the other half are connected to L. By means of this switching, reversing of polarity of the voltage divider can be achieved.

(22) A method which uses the principle of measuring inductance variance by means of U is referred to in the context of this text as a DDIS process. In accordance with the direct delta inductance sensing process (DDIS), it is assumed that the angle of the rotor relative to the stator influences the inductances of the rotor coils a, b, c, d, e and/or of the phases a, b, c, d, e. Depending on the embodiment of the motor, the cause of this effect is a change in reluctance due to the rotor-angle-dependent geometry of the magnetic circuit from the point of view of the respective motor phases or appearance of saturation in the magnetic material or both. The inductance of the coils a, b, c, d, e is produced according to the formula:

(23) $L_i\left(\right)=L_P+L_S\cos \left[2+\frac{2}{m}\left(i-1\right)\right];$ where the phase number m=5 and i=1, 2, 3, 4, 5 corresponding to the phase windings a, b, c, d, e.

(24) In this case,

(25) $L_P=\frac{L_Q+L_D}{2}$ is the component of the phase inductance which is independent of the rotor position and
L.sub.S=L.sub.QL.sub.D is the component of the phase inductance which is dependent on the rotor position.

(26) In the case where L.sub.S0, that is to say differs from the value 0, the voltage at the Y point or at the passive inverter output 111, 112, 113, 114, 115 or the passive phase terminal 111, 112, 113, 114, 115 will differ from U.sub.B/2, while the motor is in for example one of the states S1-S4. In the PWM mode, there are four alternating switching states S1, S2, S3, S4. These four switching states S1, S2, S3, S4 occur in each commutation cycle T.sub.1, T.sub.2, T.sub.3, T.sub.4, T.sub.5, in particular in each of the commutation cycles T.sub.1-T.sub.10, at least one or more times. A commutation cycle T.sub.1, T.sub.2, T.sub.3, T.sub.4, T.sub.5 in this case denotes the duration for which the at least one phase is switched to passive. The states T.sub.6, T.sub.7, T.sub.8, T.sub.9, T.sub.10 substantially correspond to the states T.sub.1, T.sub.2, T.sub.3, T.sub.4, T.sub.5 but have a different polarity and/or correspond to a half turn.

(27) Whereas in the case of three phases, it is determined which terminals are connected to the supply voltage and which are connected to the reference potential to form a voltage divider, a plurality of combinations can occur in the case of five phases. It would be possible for only one of the four active phases to be connected to the supply potential and the three others to be connected to the reference potential.

(28) In addition to commutation intervals which occur when carrying out a commutation process, the commutation cycles T.sub.1, T.sub.2, T.sub.3, T.sub.4, T.sub.5 can also be intervals during a start interval and/or detection interval. In this case, a start interval denotes an interval in which substantially no torque is generated in the rotor. An operating interval or commutation interval denotes an interval during an operating phase of the rotor in which a torque is generated, but said torque is briefly interrupted by a DDIS phase having a duty cycle of 50%. It is thus also possible to use an induction interval which substantially incorporates, inserts or induces a start interval in an operating interval. A torque is generated if a duty cycle of d is selected between the individual states which is substantially not equal to 50%. A torque is avoided if a duty cycle of d is selected between the individual states which is substantially equal to 50%.

(29) Instead of a duty cycle which is substantially used to describe the state of a half bridge or of two half bridges operated in opposing directions, in the polyphase case, by setting a state, a voltage vector may be generated in the coordinate system 202 with the axes a 202 and b 202. Said voltage vector can be represented by five duty cycles a,b,c,d,e of the coordinate system 201.

(30) To achieve a direct delta inductance sensing (DDIS) cycle, voltage differentials are formed from at least two measurements of reversely switched voltage dividers of the coils b, c, d, e connected in parallel. In a first state S1, by means of low switches 105, 105, the parallel connection of two active phases e, d is formed between the passive phase a, 141 or the branch point Y and the reference potential. In the state S1, a parallel connection of the two active phase coils d, e is correspondingly formed between the supply terminal 102 and the passive phase a, 141 or the branch point Y. In another state S3, which belongs to the voltage divider from the state S1, by means of low switches 105, 105, the parallel connection of the corresponding remaining active phases b, c is formed between the passive phase a or the branch point Y and the reference potential. In the state S3, a parallel connection of the phase coils d, e is formed between the supply terminal 102 and the passive phase 141 or the branch point Y. In other words, the switches 104, 104, 104, 104, 104, 105, 105, 105, 105, 105 are actuated in such a way that all the active phase coils are involved in forming a voltage divider between the high terminal 102 and the low terminal 103. The voltage divider has a passive phase, wherein the number of phase coils a, b, c, d which, in a switching state S1, S3, are connected between the high terminal 102 and the passive phase or the branch point Y, are equal to the number of phase coils a, b, c, d which are connected between the low connection 103 and the passive phase or the branch point Y. Correspondingly, the switching states S2, S4 may be formed from combinations of the coil pair c, d and the coil pair b, e.

(31) The phase coils a, b, c, d, e and/or phase terminals 111, 112, 113, 114, 115 which, in one switching state, are connected between the branch point Y and the low terminal 103, may be referred to as switched to low. The phase coils a, b, c, d, e and/or phase terminals 111, 112, 113, 114, 115 which, in one switching state, are connected between the branch point Y and the high terminal 102, may be referred to as switched to high. In each switching state, an inductance relation between the phase coils a, b, c, d, e is measured in the form of a voltage, depending on which phase is currently passive.

(32) When applying an alternating voltage, a change in current results from the parallel connection of the low coils b, c, and/or when generating a PWM, a total inductance results from the self-inductances of the phase coils involved in the parallel circuit. For example, in the example in the state S1 from Table 1, the total inductance L.sub.H1 of the inductances directly connected to the high terminal resulting in the switching state S1 for the parallel circuit of the high coils b, c is:

(33) $L_{H1}=\frac{L_b{L}_c}{L_b+L_c}=\frac{L_2{L}_3}{L_2+L_3}$

(34) The total inductance L.sub.L1 switched to low resulting for the parallel connection of the low coils d, e in the switching state S1 is:

(35) $L_{L1}=\frac{L_d{L}_e}{L_d+L_e}=\frac{L_4{L}_5}{L_4+L_5}$

(36) In a similar manner, in the state S2, for the parallel connection of the high coils c, d, the total inductance L.sub.H2 switched to high of the inductances directly connected to the high terminal is produced:

(37) $L_{H2}=\frac{L_c{L}_d}{L_c+L_d}=\frac{L_3{L}_4}{L_3+L_4}$

(38) The total inductance L.sub.L2 switched to low resulting for the parallel connection of the low coils b, e in the switching state S2 is:

(39) $L_{L2}=\frac{L_b{L}_e}{L_b+L_b}=\frac{L_2{L}_5}{L_2+L_5}$

(40) L.sub.H1 is the total inductance of the parallel connection switched to low in the state S3. L.sub.H2 is the total inductance of the parallel connection switched to low in the state S4. In each of the states S1, S2, S3, S4, in the phase switched to passive, an induced voltage or a voltage U.sub.1, U.sub.2, U.sub.3 and U.sub.4 caused by the self-induction of the parallel connection switched to low can be determined.

(41) The differences in voltage between U.sub.1 and U.sub.3 is U.sub.1 or U.sub.1a, U.sub.1b, U.sub.1c, U.sub.1d, U.sub.1e and/or is U.sub.2a, U.sub.2b, U.sub.2c, U.sub.2d, U.sub.2e depending on the passive phase a, b, c, d, e in which or the corresponding measurement input 120, 121, 122, 123, 124 at which measurement has been carried out. U.sub.1 is calculated following the formula

(42) 0$U_1=\frac{L_{L1}}{L_{H1}+L_{L1}}{U}_B$$U_3=\frac{L_{H1}}{L_{H1}+L_{L1}}{U}_B$$U_1=U_1-U_3=\frac{L_{L1}-L_{H1}}{L_{H1}+L_{L1}}{U}_B$

(43) U.sub.2 is calculated following the formula

(44) $U_2=\frac{L_{L2}}{L_{H2}+L_{L2}}{U}_B$$U_4=\frac{L_{H2}}{L_{H2}+L_{L2}}{U}_B$$U_2=U_2-U_4=\frac{L_{L2}-L_{H2}}{L_{H2}+L_{L2}}{U}_B$

(45) In this case, Table 1 shows the commutation duration T1=S1+S2+S3+S4 during a measurement cycle, for example without torque. In operation intermediate states may occur in addition to states S1, S2, S3, S4 to generate the required torque. During the duration T1, the phase a is switched to passive, which is shown by an O, and the phases b, c are switched to high in the switching state S1, and the phases d, e are switched to low, i.e. are connected to the high terminal and/or low terminal. In the subsequent switching state S2, the phase b and e is switched to low, and/or the phase c, d is switched to high. In the subsequent switching state S3, which belongs to S1, the phase b and c is switched to low, and/or the phase d, e is switched to high. In the subsequent switching state S4, which belongs to S2, the phase c and d is switched to low, and/or the phase b, e is switched to high. These switching phases last and are executed alternately until switching over to a new constellation at the commutation point in time at the boundary of T1 to T2 for the duration T2, in which constellation the phase b is switched to passive, and combinations of the phases a, c, d, e are switched through. After a third commutation point in time at the boundary between T2 and T3, during the duration T3, the phase c, 113 is switched to passive and switched back and forth between the phases a, b, d, e. Thus, at the boundary between T3 and T4, during the duration T4, the phase d, 114 is switched to passive, and at the boundary between T4 and T5, for the duration T5, the phase e, 115 is switched to passive. The process then repeats periodically starting with S1.

(46) In order to be able to influence the start-up of a motor M, 140 at standstill, the actuation of a motor M, 140 rotating at a low speed, or the braking of a motor M, 140, the current position of the motor M or a motor parameter is determined by an actuation sequence of the active switches 104, 104, 104, 104, 105, 105, 105, 105, impressing a torque being substantially dispensed with by selecting the duty cycle and/or voltage vector. It is thus possible for example, when the supply voltage U.sub.B is applied between the supply terminals 102, 103, to control the switches 104, 104, 104, 104, 105, 105, 105, 105 of the active bridge branches 126, 127, 128, 129 by means of the control unit 109 in such a way that, during a first period T.sub.1 the first phase terminal 111, a is switched to passive. During this period T1, the parallel connection of the second 112, b and third 113, c phase terminals and the parallel connection of the fourth 114, d and fifth 115, d phase terminals are set. This is connected alternately by the parallel connection of the third 113, c and fourth phase terminals 114, d and the parallel connection of the second 112, b and fifth 115, e phase terminals in a predeterminable duty cycle to the high terminal 102 and the low terminal 103, reversing the polarity of the formed voltage dividers also occurring between the connection.

(47) In other words, this may mean that, whilst one of the terminals 111, 112, 113, 114, 115 is switched to passive, a voltage is induced in said terminal. In addition, the phases 141, 142, 143, 144, 145 are arranged in such a way that there is substantially no magnetic and/or transformer coupling between the motor phases, or said coupling is not relevant. This may mean that a change in current in a motor phase 141, 142, 143, 144, 145 does not lead to a change in voltage in another motor phase via magnetic coupling. Only the position of the rotor relative to the stator substantially influences the magnetic flux through the phase windings and thus the impedance of the phases, which can be measured by means of the voltage differential.

(48) A rotating motor can effect two types of induction. An EMF and a change in impedance, in particular a change in the inductance and/or inductivity. The change in voltage as a result of EMF or as a result of a transformer-based effect, that is to say mutually penetrating of the coils by magnetic fields, is substantially avoided by magnetic insulation. Whereas these effects are substantially avoided, the change in the self-inductance is determined and evaluated by measuring the voltage differential U1 and/or U2. Since the voltage (EMF), which is brought about by the movement of the rotor, is substantially avoided or eliminated, the voltage which is effected by a change in impedance or induction is evaluated. In order to measure the latter voltage, the bridge branches 126, 127, 128, 129 switched to active are excited alternately by means of a PWM process in order to bring about a change in current. This alternating excitation is achieved in that the corresponding active phase terminals 112, 113, 114, 115 are alternately connected to the high terminal 102 and the low terminal 103, to which a DC voltage is connected. By alternately switching the switches on and off, the DC voltage is chopped, and in the passive phase, a voltage is induced or impressed which allows a statement about the relative position of the rotor to the stator.

(49) The period T.sub.1, during which the first phase terminal 111 is switched to passive and the four other phase terminals are switched to active corresponds to a commutation period. The commutation period is the time T.sub.1 until pole windings a, b, c, d, e have to be switched to maintain a rotational movement. These times substantially correspond to angles which are each 36. In the case of seven phases, the angles are 360/7 or 360/14. In the case of eleven phases, the angles are 360/11 or 360/22. After passing through the angle assigned to the period T1, the second phase terminal 112 takes on the role of the passive phase terminal for a time T.sub.2, then for the period T.sub.3, the third phase terminal 113, then for the period T.sub.4 the fourth phase terminal 114, and then for the period T.sub.5, the fifth phase terminal 115. The sum of the periods T.sub.1, T.sub.2, T.sub.3, T.sub.4 and T.sub.5 corresponds to the time for a half turn of an electric motor belonging to the motor M, 140.

(50) During the periods T.sub.1, T.sub.2, T.sub.3, T.sub.4 and T.sub.5, which take place between the switching of the passive phase terminals, the phases which are each switched to active are operated alternately. During this alternating active operation, in the case of a five-phase motor, firstly two of the high switches 104, 104, 104, 104 are closed and connect the corresponding phase terminals 112, 113, 114, 115 to the high terminal 102. The low switches 105, 105, 105, 105 of the bridge branches, which low switches belong to the closed high switches, behave in the exact opposite manner to the high switches of these bridge branches and are open. Thus, a voltage divider comprising coils connected in parallel is produced. The duration for which the first and second active high switches are closed is S1, the duration for which the second and third active high switches are closed is S2, the duration for which the third and fourth active high switches are closed is S3, the duration for which the fourth and first active high switches are closed is S4. Thereafter, the process repeats periodically during the corresponding periods T.sub.1, T.sub.2, T.sub.3, T.sub.4, T.sub.5.

(51) The duration of a PWM period is S1+S2+S3+S4+SX. In this case, SX describes any other states or intermediate states. This applies in particular to the special case where S1 and S2 are of the same length. In the other cases, the voltage vector is selected in such a way that no torque is generated in the corresponding measurement cycle. The duty cycle is set by means of a voltage vector in the coordinate system 202, but ambiguities can result for S1-S4-SX, i.e. for S1 to S4 and for S4 to SX. The voltage vector is standardised. In addition to the five-phase arrangement, this also applies for seven and eleven phases.

(52) Regardless of how the duty cycle is set, when selecting a duty cycle of d=50%, in the passive phase and the corresponding passive phase terminal, a voltage is induced, but no torque is generated. Therefore, selecting a duty cycle of 50% can be used to determine a motor parameter without influencing the rotational movement of the motor. During S1, the voltage U.sub.1 can be determined in the passive phase terminal, during S2, the voltage U.sub.2 can be determined in the passive phase terminal, during S3, the voltage U.sub.3 can be determined in the passive phase terminal, and during S4, the voltage U.sub.4 can be determined in the passive phase terminal. By means of these values, the two voltage differentials U.sub.1 and U.sub.2 are determined.

(53) From the two voltage differentials U.sub.1 and U.sub.2, an angle can be determined already during the time period T1, in which angle currently a rotor is arranged relative to a stator of the motor M.

(54) In order to illustrate the principle of determining angular position, FIG. 2 shows for example the rotor coordinate system 201 in relation to a stator coordinate system 202 according to an exemplary embodiment of the present invention. The rotor coordinate system 201 moves together with a rotor of the motor M in relation to the stator coordinate system 202. The abscissa , 202 and the ordinate , 202 are arranged perpendicular relative to one another and are based on a stator of the motor M and are thus stationary. The stator coordinate system 202 can be used to represent a generated total voltage vector having two coordinates a, b, the total voltage vector having been generated by setting the vector components a, b, c, d, e in the coordinate system 201. The coordinate system 201 shows five vectors a, 201, b, 201, c, 201, d, 201 and e, 201. The vectors correspond to the orientation of magnetic fields and/or the voltage components which are generated by the phase windings a, 141, b, 142, c, 143, d, 144 and e, 145 of the motor M. The vectors 201, 201, 201, 201, 201 thus correspond to a three-dimensional arrangement of the phase windings a, b, c, d, e. The phase windings a, b, c, d, e and the corresponding vectors of the magnetic fields thereof are oriented at an angle of 72 relative to one another.

(55) In one example, only the states S1 and S3 can be passed through alternately in a commutation cycle 401, T.sub.1, T.sub.2, T.sub.3, T.sub.4, T.sub.5 according to FIG. 7. Although in FIG. 7, only the two states S1 and S3 are passed through, FIG. 7 can also be used to illustrate an example in which all the states S1 to S4 are passed through in each commutation cycle T.sub.1, T.sub.2, T.sub.3, T.sub.4, T.sub.5.

(56) In other words, a voltage vector is to be set with a specific direction and amplitude. When setting the voltage vector, intermediate states can also occur. A voltage vector is described, with the components and , according to FIG. 2 in the two-dimensional coordinate system 202 for a simple illustration. However, the voltage vector is composed of components a, b, c, d, e of the coordinate system 201.

(57) Due to the fact that a motor M, having five phases a, b, c, d, e, has corresponding coils, each coil is electrically commutated after a half turn. A ten-stage commutation operation having ten commutations thus results for the five-phase motor M. The respective four active motor phases, i.e. the phases a, b, c, d, e, which are connected to the active bridge branches 125, 126, 127, 128, 129, are operated by using bipolar PWM (pulse width modulation). Due to the connection of the five phases 141, 142, 143, 144, 145 to the Y point or star point, when the bridge switches are accordingly operated against one another, four phases 141, 142, 143, 144, 145 or phase windings a, b, c, d, e are always active whilst one of the phases is switched to passive. During a first switching state S1, two high switches of two active bridge branches and two low switches of the other active bridge branches are switched on. During the first switching state S1, an electric circuit can be formed from the high terminal 102, via the two switched-on high switches 104, 104, the phase terminals 112, 113 of the motor phases b, c, the Y branch point, the phase terminals 114, 115 of the motor phases d, e, via the low switches 105, 105 to the low terminal 103. In the case of substantially the same type of construction of the four active phase coils b, c, d, e, for example b, c, d and e, an inductive voltage divider is formed, in which in each case two high phases b, c and two low phases d, e are connected in parallel. The Y branch point has a potential of U.sub.B/2 in the case, as assumed in FIG. 1, that the inductances and/or impedances of the coils of the phases are balanced.

(58) The motor control system 130 is formed by the actuating apparatus 100 or MCU (motor control unit) 100 together with the control device 109, the microcontroller 109 or the processor 109 and the sampling device 107 or measuring device 107 together with the motor M, 140.

(59) FIG. 3 shows the progression of two voltage differentials in a passive phase terminal as a function of the rotor angle, for example during forced or manual rotation of the rotor, according to an exemplary embodiment of the present invention. FIG. 3 shows a progression of two voltage differentials U.sub.1 and U.sub.1 based on the supply voltage U.sub.B (U1/U.sub.B, U2/U.sub.B) measured in the phase a, 141 switched to passive and/or in the star point Y according to an exemplary embodiment of the present invention. FIG. 3 is a graph 300 showing the two curves 303, 304 of the ratios of the voltage differentials based on the supply voltage. The progression of the differential voltage U.sub.1 is shown as the standardised ratio U1/U.sub.B in the curve 303. The progression of the differential voltage U.sub.2 is shown as the standardised ratio U2/U.sub.B in the curve 304.

(60) The curves 303, 304 were produced in that the switching patterns S1, S2, S3, S4 were applied periodically to a five-phase motor with a duty cycle of d=50% so that no torque was impressed in the motor by the current. In particular, a voltage vector may be generated in such a way that no torque is impressed, and the motor does not rotate as a result of applying a switching pattern. However, the motor was mechanically rotated to obtain measurement values for all angle values of a mechanical rotation. In this case, an external force was impressed, that is to say a torque was impressed mechanically. The active phases are switched alternately in the rhythm S1, S2, S3 and S4 while the rotor was rotated mechanically. The mechanical rotation is necessary because the duty cycle 50% is selected, and thus no torque is impressed by the active phases. The angle values of the rotor from 0 to 2 are shown on the abscissa 301. The ratios of the voltage differentials U/U.sub.B are entered on the ordinate 302 in the range from 0.15 to +0.15. The progressions of the voltage differentials 303, 304 give an unambiguous pattern by means of which an angle can be unambiguously deduced. The voltage differential curves 303, 304 can be recorded in a memory. However, they can also be determined numerically or analytically by evaluating the formulae of the inductance of the coils a, b, c, d, e and the voltage differentials. In this case, the curves can be determined by carrying out a start-up or motor recognition sequence. A start-up sequence of this type may be designed in such a way that, despite passing through a sequence S1, S2, S3, S4, no torque is generated. For example, each of the five phases can be switched to passive one after the other, and the sequence S1, S2, S3, S4 can be passed through so that five interpolation points in each case can be determined on the curve U1 303 and/or U2 304. The interpolation points are each determined at a fixed angle, since the motor does not rotate due to the lack of torque. These interpolation points can then be assumed to each lie on a curve 303, 304 and the curve shapes 303, 304 can be constructed by numerical evaluation of the formulae for the inductance and the voltage differential.

(61) After the four switching states S1, S2, S3 and S4 have been passed through, the voltages U1, U2, U3 and U4 are certain throughout the self-induction of the coils and the formation of the voltage divider. By applying connections according to Table 1, the voltages U1 and U3 belong to the same voltage divider which has the same parallel connections of phases and which is just operated with reversed polarity in the switching states S1, S3. The voltages U2 and U4 likewise belong to the same voltage divider which has the same parallel connections of phases and is just operated with reversed polarity in the switching states S2, S4. The structure of the voltage dividers, which are involved in forming U.sub.1 and U.sub.2, differs. By means of this configuration, the voltage differential U.sub.1 and U.sub.2 which is dependent on the angle of rotation can be determined according to the angle of rotation. The voltage differentials U.sub.1 and U.sub.2 may only be measured when the phases involved in forming the voltage dividers are actuated alternately, in order to generate an alternating current which produces a voltage according to the angle-dependent inductance, in particular the angle-dependent total inductance of the parallel connection.

(62) From the determined curve, the rotary position of a rotor can be determined during operation and transferred to a commutation process. As a commutation condition, two U values are determined which are used as a threshold to carry out a commutation when the thresholds are exceeded. These commutation conditions are determined while the motor is free from torque and monitored during the movement of the motor in order to be able to commutate at the right time. In order to determine the curves 303, 304 or in order to monitor the commutation condition, a lookup table of the curves 303, 304 can be recorded in the memory of the controller 109. In each commutation step, however, other U can be relevant as a commutation condition, i.e. in each case 5 U per half turn. After a half turn, the values repeat.

(63) Motor parameters can also be determined by voltage differential curves. Said motor parameter can be transferred to the commutation process. In order to deduce the rotary position from the graph 300, during at least one pass through the sequence S1, S2, S3, S4, the current standardised ratio U1/U.sub.B and U2/U.sub.B is detected, and the angle belonging to the two values is read on the abscissa 301. In other words, a voltage pattern is detected in such a way that the currently measured values ratio U1/U.sub.B and U2/U.sub.B can be assigned to precisely one angle. This angle which can be assigned to both values indicates the current rotary position of the rotor. This determination can be carried out when at a standstill in a start interval or also during an operating interval. The minimum time for a determination is the sum of the durations for S1, S2, S3, S4, for example T1=S1+S2+S3+S4. For this purpose, however, the curves 303, 304 should already be present in the memory to avoid inaccuracies. In one example, T1 can be the commutation time and have a duration which corresponds to a multiple of the minimum duration of S1+S2+S3+S4. The values of the curves 303, 304 can be stored as curves or as tables to allow angle detection. The recognition of the voltage pattern and the determination of the corresponding angle can take place either visually, by visually comparing the curves, and/or numerically in that the measurement values are visually searched for on the curves and/or in that values are retrieved from the table. In order to increase the precision, average values can be formed from a plurality of measurements and/or additional passive phases can be incorporated. For example, the times T2, T3, T4, T5 can indicate periods during which the corresponding phase b, c, d, e is switched to passive.

(64) The curves 303, 304 are sinus-like and are phase-shifted by approximately 90. Due to the selected combination of phase coils to voltage dividers, the amplitude of the curve 303 is higher than the curve 304. This increased amplitude results from the application of the formulae for the inductance and the voltage differentials. In order to determine the curve 303, the voltage across the parallel connection of the phase coils d, e and b, c is evaluated alternately. In order to determine the curve 304, the voltage across the parallel connection of the phase coils b, e and c, d is evaluated alternately.

(65) The motor M, 140 is operated in a ten-step commutation mode to create the curves 303, 304. The four active motor phases are operated to generate a voltage vector. By means of the coordinate systems 202 a, b, a composed voltage vector of the components b, c, d, e, 201, 201, 201, 201 can be formed in T1. This voltage vector effects a current vector which in turn, depending on the rotor position, effects a momentum in a specific direction. The voltage is measured at least in switching states in which two phases are connected to the supply voltage, and the two other voltages are connected to the reference potential. This results in four states S1, S2, S3, S4 for the five-phase machine. At least one measurement of the states at two different points in time is used to pass through a DDIS cycle. Table 1 shows the switching states in which voltage measurements were carried out and in which the phase a is passive. The ratios U1/U.sub.B and U2/U.sub.B are measured at the branch point Y and/or at the phase a. The phase measurement can be carried out when the individual phases are magnetically insulated from one another so that no coupling of the magnetic fields of the coils to one another takes place.

(66) The passive phase in which the curve shape from the graph 300 is produced can be used as an angle sensor (resolver) to determine the angle of rotation. The current through the active phases can be different in size. In the case of a three-phase motor, the current through the active phases is substantially always the same, since there are no further branching options for the current than through the two active phases connected in series. In the case of the five-phase motor, due to the parallel connection of at least two active phases to form a voltage divider branch, an asymmetrical current distribution can occur in the parallel connection. In the case of parallel connection of the phase branches, the current can be distributed asymmetrically on the phase branches. This current distribution can be brought about in a targeted manner by the selection of the voltage vector or can result from the voltage vector being currently applied.

(67) FIG. 4 shows the progression 401 of an absolute voltage in a passive phase a in a simulation with a corresponding circuit diagram 402 of the switching states according to one exemplary embodiment of the present invention. The switching states are marked out on the x-axis 403, 404 as a function of time t. The voltage values U.sub.1,2,3,4 are marked out on the ordinate 406 and range from 0 to 12. In graph 402, the four switching diagrams of the active phases b, c, d, e and in particular the high switches and low switches thereof are shown in the form of phase-shifted rectangular curves. The rectangular curves switch back and forth between the state H and L and are mutually phase-shifted in such a way that the states S1, S2, S3, S4 are produced periodically. Each of the rectangular curves per se forms a PWM, in particular a bipolar PWM, and generates a change in current in the corresponding phase. The state H means that the corresponding phase is connected to the supply voltage U.sub.B by means of the high switch, and the state L means that the corresponding phase is connected to the reference potential by means of the low switch. On the ordinate 405, the switching states H, L by means of which the switches of the actuating apparatus 100 can be controlled are indicated.

(68) From FIG. 4, it can be seen that graph 402 starts with the state S4, in which the phase b receives the control pulse H, the phase c receives the control pulse L, the phase d receives the control pulse L, and the phase e receives the control pulse H. The state S1 follows, in which the phase b receives the control pulse H, the phase c receives the control pulse H, the phase d receives the control pulse L, and the phase e receives the control pulse L. The switching states S1, S2 and S3 follow. It is to be seen that an active phase is kept constant on average over two clock states. After passing through the states S4, S1, S2, S3, the sequence repeats periodically in the order S4, S1, S2, S3. Whilst the control pulses 402 are generated periodically by the control device 109, the rotor is mechanically rotated in a uniform manner. The rotation is mechanically forced. The duty cycle of the control sequence 402 is selected in such a way that no torque is generated by the supplied electrical power, as a result of which the mechanical rotation is necessary for movement.

(69) In the curve 401, the voltage progression is shown which results when applying the actuation pattern according the control curves 402 in the phase a switched to passive. The applied control sequence 402 is configured in such a way that a modulated voltage vector which is zero results. If a voltage vector not equal to zero is applied with a corresponding PWM pattern, there are more switching states than S1, S2, S3, S4, and even intermediate states can occur. In the case of an intermediate state, for example when effecting a switching pattern, a state can accrue in which for example three phase connections of the four active phase terminals are switched to H and only one is switched to L, or in which three phase terminals are switched to L and only one is switched to H.

(70) In the time range from 0.0025 s to 0.0029 s on the time axis 404, a region A is indicated. This leads to the sub-segment 410 of the voltage progression 401 in the graph 401. In this region, the voltages U1 and U3 effected by the switching states S1 and S3 are approximately the same. Thus, in the region 410, the voltage differential curve U1/U.sub.B, 303 is equal to zero and has a zero-crossing. This zero-crossing 410 is also marked in the curve 303 in FIG. 3 and is dependent on the angle of rotation.

(71) In the time range from 0.0032 to 0.0036 on the time axis 404, a region B is indicated. The region B is phase-shifted by 90 relative to the region A. This region B leads to the sub-segment 411 of the voltage progression 401 in the graph 401. In this region B, the voltages U2 and U4 effected by the switching states S2 and S4 are approximately the same. Thus, in the region 411, the voltage differential curve U2/U.sub.B, 304 is equal to zero and has a zero-crossing. This zero-crossing 411 is also marked in the curve 304 in FIG. 3 and is phase shifted by about 90. By means of the phase shift, the voltage patterns which are to be found in order to determine the angles, are separated very clearly from one another as a result of which the angles of rotation by means of this phase shift can be well recognised. The 90 phase shift results from the actuation, as can be seen for example from FIG. 2. Thus, for example in state S1 phases b and c are on and a vector parallel to axis results. The ability of well recognising the angles with the selected curves 303, 304 exists primarily due to the fact that two curves are available and there is always a curve with a sharp increase per angle, as in the case of a resolver.

(72) FIG. 5 shows section A from FIG. 4 in proximity to the zero-crossing of the voltage differential curve U1/U.sub.B, 303 according to one embodiment of the present invention. In this enlarged view, the switching states S1, S2, S3, S4 and the resulting voltages U.sub.1, U.sub.2, U.sub.3, U.sub.4 can be seen. In the region of the zero-crossing 410, U.sub.1 and U.sub.3 are approximately equal, and therefore a voltage differential U1/U.sub.B=0 is produced.

(73) FIG. 6 shows a section from FIG. 4, which is phase shifted by 90 relative to FIG. 5 according to an exemplary embodiment of the present invention. In particular FIG. 6 shows a section from FIG. 4 in which the rotor is rotated further by 90 relative to FIG. 5. FIG. 6 shows the section B from FIG. 4 in proximity of the zero-crossing of the voltage differential curve U2/U.sub.B, 304. In this enlarged view, the periodically repeating switching states S1, S2, S3, S4 and the resulting voltages U.sub.1, U.sub.2, U.sub.3, U.sub.4 can be seen. In the region of the zero-crossing 411, U.sub.2 and U.sub.4 are approximately equal, and therefore a voltage differential U2/U.sub.B=0 is produced.

(74) FIG. 7 shows the sequence of various switching states according to an exemplary embodiment of the present invention. One timing chart 700a, 700b, 700c, 700d, 700e is indicated per phase a, b, c, d, e. The abscissa 703 of each of the charts indicates the time range which ranges from 0.0000 s to 0.0030 s. Although in FIG. 7, only the two states S1 and S3 are passed through, FIG. 7 can also be used to illustrate an example in which all the states S1 to S4 are passed through in each commutation cycle T.sub.1, T.sub.2, T.sub.3, T.sub.4, T.sub.5.

(75) In chart 700a, the actuation of the phase a is shown. It can be seen that the voltage progression 701a regularly alternates between H and L to generate a PWM. This PWM is interrupted by the commutation intervals T1. During these commutation intervals, the phase a is switched to passive. The passive switching can be recognised, since the current through this phase a, which current is indicated by the current curve 702a, decreases to the value zero during T1. During the time T1, the induced absolute voltage progression 401 can be measured in the passive phase a, from which progression the progressions of the voltage differential ratios U1/U.sub.B and U2/U.sub.B can be determined. At the end of the time interval T1, at the commutation boundary, the phase a is required to continue to drive the motor. The phase contributes to the current. The phase a is switched to active, and the interval T2 follows, as can be seen from the chart 700c.

(76) To additionally determine the angle of rotation, the next phase coil is switched to passive, in the example of FIG. 7, the phase c. In chart 700c, it can be seen that the voltage progression 701c regularly alternates between H and L to generate a PWM. At the interval boundaries, phase shifting can occur during operation of the individual high switches or low switches relative to the preceding interval. To generate a torque, a vector is generated in that a suitable pattern is calculated. However, all the active phases are always used, for example by the pattern S1-S4. This PWM is interrupted by the commutation intervals T2. During these commutation intervals T2, the phase c is switched to passive. The passive switching can be recognized, since the current in the phase c, which current is indicated by the current curve 702c, decreases to the value zero during T2. During the time T2, the induced absolute voltage progression can be measured in the passive phase c, from which the progressions of the voltage differential ratios U1/U.sub.B and U2/U.sub.B can be determined. At the end of the time interval T2, at the commutation boundary, the phase c is required to continue to drive the motor. The phase c is switched to active, and the interval T3 follows, during which the phase e is switched to passive.

(77) To additionally determine the angle of rotation, the next phase coil is switched to passive, in the example of FIG. 7, the phase e, is switched to passive. In chart 700b, it can be recognised that the voltage progression 701e regularly alternates between H and L to generate a PWM. This PWM is interrupted by the commutation intervals T3. During these commutation intervals T3, the phase e is switched to passive. The passive switching can be recognised, since the current in the phase e decreases to zero during T3, which current is indicated by the current curve 702e. During the time T3, the induced absolute voltage progression can be measured in the passive phase e, from which voltage progression the progressions of the voltage differential ratios U1/U.sub.B and U2/U.sub.B can be determined. At the end of the time interval T3, at the commutation boundary, the phase e is required to continue to drive the motor. The phase e is switched to active, and the interval T4 follows, during which the phase b is switched to passive.

(78) To additionally determine the angle of rotation, the next phase coil is switched to passive, in the example of FIG. 7, the phase b, is switched to passive. In chart 700b, it can be recognised that the voltage progression 701b regularly alternates between H and L to generate a PWM. This PWM is interrupted by the commutation intervals T4. During these commutation intervals T4, the phase b is switched to passive. The passive switching can be recognised, since the current in the phase b decreases to the value zero during T4, which current is indicated by the current curve 702b. During the time T4, the induced absolute voltage progression can be measured in the passive phase b, from which voltage progression the progressions of the voltage differential ratios U1/U.sub.B and U2/U.sub.B can be determined. At the end of the time interval T4, at the commutation boundary, the phase b is required to continue to drive the motor. The phase b is switched to active, and the interval T5 follows, during which the phase d is switched to passive.

(79) To additionally determine the angle of rotation, the next phase coil is switched to passive, in the example of FIG. 7, the phase d, is switched to passive. In chart 700d, it can be recognised that the voltage shape 701d regularly alternates between H and L to generate a PWM. This PWM is interrupted by the commutation intervals T5. During these commutation intervals T5, the phase d is switched to passive. The passive switching can be recognised, since the current in the phase d decreases to the value zero during T5, which current is indicated by the current curve 702d. During the time T5, the induced absolute voltage progression can be measured in the passive phase d, from which voltage progression the progressions of the voltage differential ratios U1/U.sub.B and U2/U.sub.B can be determined. At the end of the time interval T5, at the commutation boundary, the phase d is required to continue to drive the motor. The phase d is switched to active, and the interval T1 follows again, during which the phase a is switched to passive.

(80) After passing through the sequence T1, T2, T3, T4, T5, an electric half turn is performed, i.e. an angle of rotation of 180 is covered. The full sequence repeats periodically. Only the periods T1, T2, T3, T4, T5 can be longer or shorter depending on the turning rate of the motor. They are therefore turning rate-dependent.

(81) FIG. 8 is a flow chart for a method for actuating a motor according to an exemplary embodiment of the present invention. The method for actuating a polyphase motor 140, M starts in the starting state S800.

(82) In the state S801, a supply voltage U.sub.B is applied to a high terminal 102 of an actuating apparatus 100. In the state S802, a reference potential of the supply voltage U.sub.B is applied to a low terminal 103 of an actuating apparatus 100.

(83) In the state S803, a pulse-width-modulated voltage pattern 402, 701a, 701b, 701c, 701d, 701e is impressed in four of the five phase terminals 111, 112, 113, 114, 115 of the actuating apparatus 100 by connecting the phase terminals 115, 112, 113, 114 to the high terminal 102 or the low terminal 103.

(84) In the state S804, an evaluation signal which is dependent on the angle of rotation of the polyphase motor 140, M is detected in a fifth phase terminal 111. In addition, an angle of rotation and/or a commutation condition of the polyphase motor 140, M is determined from the evaluation signal.

(85) The method ends at the end state S805.

(86) In addition, it should be noted that the terms comprising and having do not exclude any other elements or steps and a or an does not exclude a plurality. Furthermore, it should be noted that features or steps which have been described with reference to one of the above embodiments, can also be used in combination with other features or steps of other above-described embodiments. Reference numerals in the claims should not be understood as limiting.