Motor drive device

Abstract

This motor drive device comprises: a position detection unit (10) that outputs a position detection signal by detecting a counter electromotive force generated in a motor winding during a non-energization period in which the motor winding is not energized; a current waveform control unit (20) that controls a current waveform including a non-energization period; an energization control unit (30) that controls energization to the motor winding on the basis of the current waveform; and a drive unit (40) that supplies a current to the motor winding according to the control of the energization control unit (30). The energization control unit (30) switches the energization phase of the motor (100) provided with the motor winding on the basis of the position detection signal. The current waveform control unit (20) adjusts the start timing of the non-energization period on the basis of the non-energization period.

Description

Motor drive unit

[0001] The present disclosure relates to a motor drive device, and more particularly to a motor drive device for driving a sensorless motor that does not have a position detection element for detecting a rotor position.

[0002] A sensorless motor driver, which does not have a position detection element, detects the rotor rotation position by detecting the zero crossings of the back electromotive force (back EMF) that occurs as the rotor rotates, appearing as a potential difference between the energized terminal and the neutral terminal of each winding, and controls the timing of switching the energized phase based on this rotation position information. To accurately detect back EMF, a non-energized period is typically provided during back EMF detection, during which no current is passed through the motor windings, and the current flowing through the motor winding of the phase being detected is controlled to zero. However, if this non-energized period is too long, the motor current will be distorted, causing noise or vibration. Conversely, if it is too short, the zero crossings of the back EMF cannot be detected. Patent Document 1 discloses a motor driver that can change the start timing of the non-energized period.

[0003] Japanese Patent Application Laid-Open No. 2011-244617

[0004] However, in the technology disclosed in Patent Document 1, the start timing of the non-energized period is selected from predetermined candidates using a control signal corresponding to the rotational state of the motor, which results in problems such as the need for a circuit to detect the rotational state of the motor, and the limited number of options for the non-energized period making it impossible to sufficiently shorten the non-energized period.

[0005] In view of the above problems, an object of the present disclosure is to provide a motor drive device that is capable of optimizing the non-energized period.

[0006] In order to solve the above problems, the motor drive device disclosed herein is a motor drive device for driving a sensorless motor, and includes a position detection unit that outputs a position detection signal by detecting back electromotive force generated in the motor windings during a non-energization period when no current is applied to the motor windings, a current waveform control unit that controls a current waveform including the non-energization period, an energization control unit that controls energization to the motor windings based on the current waveform, and a drive unit that supplies current to the motor windings in accordance with the control of the energization control unit, wherein the energization control unit switches the energization phase of a motor that has the motor windings based on the position detection signal, and the current waveform control unit adjusts the start timing of the non-energization period based on the non-energization period.

[0007] These comprehensive or specific aspects may be realized as a system, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM, or may be realized as any combination of a system, a method, an integrated circuit, a computer program, and a recording medium.

[0008] A motor drive device according to an aspect of the present disclosure makes it possible to optimize the non-energized period.

[0009] FIG. 1 is a circuit configuration diagram showing an example of a motor drive device according to a first embodiment. FIG. 2 is a circuit configuration diagram showing an example of a position detection unit and a non-energized period designation unit according to the first embodiment. FIG. 3 is a table showing an example of adjustment of a non-energized period in the first embodiment. FIG. 4 is a waveform diagram for explaining the operation of the position detection unit according to the first embodiment. FIG. 5 is a waveform diagram for explaining the operation of the non-energized period designation unit according to the first embodiment. FIG. 6 is a circuit configuration diagram showing an example of a position detection unit and a non-energized period designation unit according to a second embodiment. FIG. 7 is a table showing an example of adjustment of a non-energized period in the second embodiment. FIG. 8 is a waveform diagram for explaining the operation of the non-energized period designation unit according to the second embodiment. FIG. 9 is a circuit configuration diagram showing an example of a non-energized period designation unit according to a third embodiment. FIG. 10 is a table showing an example of adjustment of a non-energized period in the third embodiment. FIG. 11 is a circuit configuration diagram showing an example of a non-energized period designation unit according to a fourth embodiment. FIG. 12 is a table showing an example of adjustment of a non-energized period in the fourth embodiment. FIG. 13 is a waveform diagram for explaining the operation of the non-energized period designation unit according to the fourth embodiment. 10 is a circuit diagram showing an example of a position detection unit according to a fifth embodiment; FIG. 11 is a waveform diagram for explaining the operation of a non-energized period designation unit according to another embodiment; FIG. 12 is a waveform diagram for explaining the operation of a non-energized period designation unit according to another embodiment;

[0010] First Embodiment FIG. 1 is a circuit diagram showing an example of a motor drive device according to a first embodiment. The motor drive device according to this embodiment is a drive device for driving a sensorless motor. A motor 100, which is the target of the motor drive device according to this embodiment, is, for example, a three-phase sensorless motor. In FIG. 1 , a position detector 10 detects zero crossings of the back electromotive force generated in the motor windings U, V, and W during a non-energized period when the motor windings U, V, and W are not energized, and outputs a position detection signal BEMF. Specifically, the position detector 10 detects zero crossings of the back electromotive force generated in the motor winding U during a non-energized period when the motor winding U is not energized, detects zero crossings of the back electromotive force generated in the motor winding V during a non-energized period when the motor winding V is not energized, and detects zero crossings of the back electromotive force generated in the motor winding W during a non-energized period when the motor winding W is not energized. The position detector 10 detects the drive voltages to the motor windings U, V, and W and the neutral point potential Vc of the motor 100 in order to detect the zero crossing of the back electromotive force.

[0011] The current waveform control unit 20 controls the current waveform to be supplied to the motor 100 based on the torque command signal TQ and the position detection signal BEMF. The current waveform includes a non-current-carrying period. That is, the current waveform control unit 20 controls a current waveform including a non-current-carrying period. The current waveform control unit 20 is composed of a non-current-carrying period designation unit 21 and a waveform generation unit 22. The non-current-carrying period designation unit 21 outputs a non-current-carrying period signal TP indicating the non-current-carrying period, which rises at a predetermined start timing tp and falls in accordance with the position detection signal BEMF. The non-current-carrying period designation unit 21 times the non-current-carrying period indicated by the non-current-carrying period signal TP and sets the start timing tp of the non-current-carrying period of the next cycle based on the time information measured. The waveform generation unit 22 generates a current waveform based on the torque command signal TQ and the non-current-carrying period signal TP.

[0012] The current control unit 30 controls the current supply to the motor windings based on the generated current waveform. Specifically, the current control unit 30 generates a control signal for controlling the current supply to each motor winding by performing modulation such as PWM (Pulse Width Modulation) on the current waveform, and switches the current supply phase of the motor 100 including each motor winding based on the position detection signal BEMF.

[0013] The drive unit 40 supplies current to each motor winding under the control of the energization control unit 30. The drive unit 40 has a configuration in which a high-side switch and a low-side switch, each equipped with a body diode or an anti-parallel-connected diode, are connected in series between a power supply (Vdd) and ground. The drive unit 40 alternately turns on the high-side switch and the low-side switch at a predetermined frequency during periods other than the non-energized period, and turns off both the high-side switch and the low-side switch during the non-energized period. For example, the high-side switch 41 and the low-side switch 42 corresponding to the motor winding U alternately switch on and off in accordance with a control signal from the energization control unit 30 to supply current to the motor winding U. In FIG. 1 , the high-side switch 41 and the low-side switch 42 are represented as field-effect transistors having body diodes, but the switching elements included in the motor drive device are not limited to field-effect transistors. For example, the switching elements included in the motor drive device may structurally include an anti-parallel diode such as a body diode, or may have an anti-parallel diode separately connected.

[0014] Furthermore, in this disclosure, the configuration and operation of the motor drive device will be described by focusing on the U phase of the motor windings U, V, and W, and referring to the drive voltage of the motor winding U as the phase potential Vu and the current of the motor winding U as the current Iu.

[0015] FIG. 2A is a circuit diagram showing an example of the position detection unit 10 and the non-energized period designation unit 21 according to the first embodiment.

[0016] As shown in FIG. 2A, the position detection unit 10 includes, for example, a comparator 11 and a one-shot pulse circuit 12. The comparator 11 is activated by the non-energized period signal TP. For example, when the non-energized period signal TP is at an L level, the comparator 11 also outputs an L level. The comparator 11 also compares the drive voltage (phase potential Vu) of the motor winding U with the neutral point potential Vc of the motor 100. The one-shot pulse circuit 12 outputs a one-shot pulse by detecting the falling edge of the output signal CMP of the comparator 11. The output of this one-shot pulse circuit 12 is the position detection signal BEMF.

[0017] The de-energized period designation unit 21 includes a timer circuit 210 , a comparison circuit 211 , a de-energized timing setting circuit 212 , and a latch circuit 213 .

[0018] The timing circuit 210 times the non-energized period and outputs a time signal Tp indicating the length of the non-energized period. The time signal Tp may be, for example, an analog value indicating the voltage of the capacitor due to constant current charging to the capacitor, or a digital value indicating the number of counts of a predetermined clock by a counter. Since the measured time indicated by the time signal Tp corresponds to the non-energized period, hereinafter the non-energized period will also be referred to as the non-energized period Tp.

[0019] The comparison circuit 211 compares the non-energizing period Tp with a predetermined value to determine the level, and the non-energizing timing setting circuit 212 adjusts the start timing tp of the non-energizing period Tp based on the output of the comparison circuit 211 .

[0020] FIG. 2B is a table showing an example of adjusting the non-energized period Tp in the first embodiment.

[0021] In this embodiment, the comparison circuit 211 sets a predetermined value T1 and a predetermined value T2 greater than the predetermined value T1 as predetermined values ​​to be compared with the non-energized period Tp. The predetermined value T1 is an example of a first predetermined value, and the predetermined value T2 is an example of a second predetermined value.

[0022] 2B , when the non-energization period Tp is equal to or shorter than the predetermined value T1 (Tp≦T1), both the comparison result a1 with the predetermined value T1 and the comparison result a2 with the predetermined value T2 are H level, so that the non-energization timing setting circuit 212 determines that the non-energization period Tp is too short and advances the start timing tp of the non-energization period Tp (e.g., advances the start timing tp by a change amount ΔT: tp → tp − ΔT). 2B , when the non-energization period Tp is longer than the predetermined value T1 but equal to or shorter than the predetermined value T2 (T1<Tp≦T2), the comparison result a1 with the predetermined value T1 is L level and the comparison result a2 with the predetermined value T2 is H level, so that the non-energization timing setting circuit 212 determines that the non-energization period Tp is appropriate and maintains the start timing tp of the non-energization period Tp (tp → tp). As shown in FIG. 2B , when the non-energized period Tp is longer than the predetermined value T2 (Tp>T2), both the comparison result a1 with the predetermined value T1 and the comparison result a2 with the predetermined value T2 are at the L level, so the non-energized timing setting circuit 212 determines that there is too much margin in the non-energized period Tp and delays the start timing tp of the non-energized period Tp (for example, delays the start timing tp by a change width ΔT: tp → tp + ΔT).

[0023] In this way, the current waveform control unit 20 adjusts the start timing tp of the non-energization period Tp based on the non-energization period Tp. As a result, if the non-energization period Tp, which is the period from when the supply of current to the motor windings is stopped to when a zero crossing of the back electromotive force is detected, is too long, the start timing tp of the non-energization period Tp can be adjusted to shorten the non-energization period Tp. If the non-energization period Tp is too short, the start timing tp of the non-energization period Tp can be adjusted to lengthen the non-energization period Tp. For example, instead of selecting the start timing tp of the non-energization period Tp from predetermined candidates, the start timing tp of the non-energization period Tp can be flexibly adjusted by feedback control of the non-energization period Tp. This enables optimization of the non-energization period Tp. For example, the non-energization period Tp can be shortened while still setting the non-energization period Tp necessary for rotor position detection, thereby suppressing vibration, noise, and malfunction.

[0024] For example, when the current waveform control unit 20 detects that the non-current-carrying period Tp is shorter than a predetermined value T1, it advances the start timing tp of the non-current-carrying period Tp. If the non-current-carrying period Tp is too short, it is possible to optimize the non-current-carrying period Tp by advancing the start timing tp of the non-current-carrying period Tp so that the non-current-carrying period Tp is longer.

[0025] For example, when the current waveform control unit 20 detects that the non-current-carrying period Tp is longer than a predetermined value T2, it delays the start timing tp of the non-current-carrying period Tp. If the non-current-carrying period Tp is too long, the start timing tp of the non-current-carrying period Tp is delayed so as to shorten the non-current-carrying period Tp, thereby optimizing the non-current-carrying period Tp.

[0026] The latch circuit 213 latches an H level signal in synchronization with the output of the de-energization timing setting circuit 212, and is reset by the position detection signal BEMF via an inverter 214. The Q output of the latch circuit 213 becomes the de-energization period signal TP.

[0027] The operation of the motor drive device according to this embodiment having the above configuration will be described with reference to the waveform diagrams of Figures 3 and 4. First, the operation of the position detection unit 10 will be described with reference to Figure 3.

[0028] FIG. 3 is a waveform diagram illustrating the operation of the position detector 10 according to the first embodiment. From top to bottom, FIG. 3 shows the back electromotive force Bemf, the non-energized period signal TP, the phase potential Vu and neutral point potential Vc, the current Iu, the output CMP of the comparator 11, and the position detection signal BEMF. The phase potential Vu is a high-frequency rectangular wave generated by the switching operation of the high-side switch 41 and the low-side switch 42 except during the non-energized period Tp, but is shown as a smoothed waveform in FIG. 3 . That is, if the duty ratio (the proportion of on-time in one switching cycle) of the high-side switch 41 during switching operation is δ and the power supply potential is Vdd, then Vu = δ × Vdd. Similarly, the phase potential Vu is shown as a smoothed waveform in the subsequent waveform diagrams.

[0029] The non-energized period designation unit 21 sets the non-energized period signal TP to the H level from time tp until the next pulse input of the position detection signal BEMF. This causes the waveform generation unit 22 to generate a current waveform that is zero during the H level of the non-energized period signal TP. In accordance with this current waveform, the energization control unit 30 outputs control signals that turn off both the high-side switch 41 and the low-side switch 42 during the H level of the non-energized period signal TP. When the high-side switch 41, which had been supplying current to the motor winding U, is turned off, the phase potential Vu drops sharply, the body diode of the low-side switch 42 becomes conductive, and a regenerative current flows as the current Iu. The regenerative current is a decreasing current, and during the period when the regenerative current is flowing, the phase potential Vu is lower than the ground potential by the forward voltage of the body diode of the low-side switch 42. The comparator 11 is active while the non-energized period signal TP is at H level, but the output CMP is at L level because the phase potential Vu is lower than the neutral point potential Vc.

[0030] When the regenerated current Iu becomes zero at time t1, the phase potential Vu rises and becomes higher than the neutral point potential Vc, and the output CMP of the comparator 11 becomes H level. At this time, the current Iu does not flow, so a back electromotive force Bemf superimposed on the neutral point potential Vc is generated in the phase potential Vu.

[0031] At time t2, when the phase potential Vu falls below the neutral point potential Vc, i.e., when the back electromotive force Bemf crosses zero, the output CMP of the comparator 11 falls to the L level. The one-shot pulse circuit 12 detects this falling edge and outputs the position detection signal BEMF. Following the rising edge of the position detection signal BEMF, the latch circuit 213 of the non-energized period designation unit 21 is reset, and the non-energized period signal TP falls to the L level, ending the non-energized period Tp. Following the end of the non-energized period Tp, the high-side switch 41 and low-side switch 42 of the drive unit 40 resume their switching operation in accordance with the current waveform generated by the waveform generation unit 22, and begin supplying current to the motor winding U.

[0032] Next, an example of the operation of the motor drive device of this embodiment until the long non-energized period Tp reaches an appropriate value will be described with reference to FIG.

[0033] 4 is a waveform diagram illustrating the operation of the non-energized period designation unit 21 according to the first embodiment. From the top to the bottom, FIG. 4 shows the back electromotive force Bemf, the non-energized period signal TP, the phase potential Vu, the neutral point potential Vc, and the position detection signal BEMF. The time 0 on the horizontal axis in the diagram indicates the timing at which the position detection signal BEMF rises, and the rising edge of the position detection signal BEMF resets the time information for timing the non-energized period Tp to zero.

[0034] Assume that the start timing tp of the non-energization period Tp is time tp1, and that the timing circuit 210 starts measuring the non-energization period Tp from time tp1, and the comparison circuit 211 determines that Tp > T2. In this case, the non-energization timing setting circuit 212 delays the start timing tp of the next non-energization period Tp from time tp1 by a change amount ΔT, and sets it to time tp2 (tp2 = tp1 + ΔT). If the comparison circuit 211 also determines that the non-energization period Tp measured from time tp2 of the next cycle is Tp > T2, the non-energization timing setting circuit 212 delays the start timing tp of the next non-energization period Tp by a change amount ΔT from time tp2, as in the previous cycle. Then, assume that the non-energization period Tp measured from time tp3 (= tp2 + ΔT) of the next cycle is determined to be T1 < Tp ≦ T2. In this case, the de-energization timing setting circuit 212 maintains the start timing tp of the next de-energization period Tp at time tp3, thereby maintaining the de-energization period Tp thereafter between predetermined value T1 and predetermined value T2.

[0035] As described above, according to this embodiment, the zero-cross detection period (non-energized period Tp) in sensorless motor drive can be shortened so that it approaches a target value, thereby enabling low-vibration and low-noise drive.

[0036] In the first embodiment, the predetermined values ​​T1, T2, and the variation width ΔT are set arbitrarily in design. For example, the variation width ΔT may be different when the start timing tp is delayed and when it is advanced. Furthermore, the above values ​​do not have to be fixed values ​​and may be changed as appropriate.

[0037] Furthermore, the change in the start timing of the non-power-on period Tp does not have to be determined solely by comparing one non-power-on period Tp with a predetermined value, but may be determined by the fact that the comparison results are the same for a predetermined number of consecutive periods, or may be determined by the average value of the comparison results for a predetermined number of periods.

[0038] For example, if the start timing tp of the non-current-carrying period Tp is advanced each time it is detected that the non-current-carrying period Tp is shorter than the predetermined value T1, the non-current-carrying period Tp will be changed frequently. Therefore, the current waveform control unit 20 may advance the start timing tp of the non-current-carrying period Tp when it detects that the non-current-carrying period Tp is shorter than the predetermined value T1 a predetermined number of times. This makes it possible to suppress vibrations, noise, or malfunctions that accompany frequent changes to the non-current-carrying period Tp.

[0039] For example, if the start timing tp of the non-power-conduction period Tp is delayed every time it is detected that the non-power-conduction period Tp is longer than the predetermined value T2, the non-power-conduction period Tp will be changed frequently. Therefore, the current waveform control unit 20 may delay the start timing tp of the non-power-conduction period Tp when it detects that the non-power-conduction period Tp is longer than the predetermined value T2 a predetermined number of times. This makes it possible to suppress vibration, noise, or malfunctions that accompany frequent changes to the non-power-conduction period Tp.

[0040] Second Embodiment In the first embodiment, an example was described in which the non-power-supply period Tp, which is set as a detection period for detecting zero crossings of the back electromotive force, is adjusted toward a target value. In contrast, in the second embodiment, an example is described in which a period obtained by excluding a regeneration period from the non-power-supply period Tp is set as a detection period for detecting zero crossings (zero crossing detection period), and the non-power-supply period Tp is adjusted so as to shorten this detection period to a target value.

[0041] 5A is a circuit diagram showing an example of a position detection unit 10A and a non-energized period designation unit 21A according to the second embodiment. Note that other configurations are the same as those in the first embodiment, and therefore illustrations and descriptions thereof will be omitted.

[0042] In the second embodiment, the position detector 10A generates a detection period signal ENB indicating a detection period, which is a portion of the non-energized period Tp, from the end of regenerative operation in which current flows through the motor windings until the generation of the position detection signal BEMF. Specifically, the position detector 10A detects zero crossings of the back electromotive force generated in the motor windings during the non-energized period Tp to output the position detection signal BEMF and the detection period signal ENB. As shown in FIG. 5A , the position detector 10A includes, for example, a comparator 11A, a comparator 13, and an AND circuit 14. The comparator 11A is activated by the detection period signal ENB. For example, when the detection period signal ENB is at a low level, the comparator 11A also outputs a low level. The comparator 11A also compares the phase potential Vu with the neutral point potential Vc. The phase potential Vu is also compared with ground potential by the comparator 13, and when the phase potential Vu falls below ground potential, the comparator 13 outputs a signal CMP1 that goes low. The AND circuit 14 outputs a detection period signal ENB when it receives the signal CMP1 and the non-energized period signal TP. As described above, the position detector 10A has a comparison circuit (comparator 13) that compares the potential at the junction between the high-side switch 41 and the low-side switch 42 (phase potential Vu) with ground potential, and generates the detection period signal ENB during the non-energized period Tp, which is a period when the potential at the junction (phase potential Vu) is higher than ground potential, as a detection period.

[0043] The non-energized period designation unit 21A has a timer circuit 210, a comparison circuit 211A, a non-energized timing setting circuit 212A, and a latch circuit 213. The non-energized period designation unit 21A measures the time from when the detection period signal ENB rises to when the position detection signal BEMF rises, and outputs a non-energized period signal TP for the next cycle based on the measured time information. Note that the time information is reset by the position detection signal BEMF.

[0044] The timing circuit 210 measures the time from when the detection period signal ENB rises to when the position detection signal BEMF rises, and outputs a time signal DT indicating the measured time information. The time signal DT may be, for example, an analog value indicating the voltage of the capacitor due to constant current charging of the capacitor, or a digital value indicating the number of counts of a predetermined clock by a counter. The measured time indicated by the time signal DT corresponds to the zero-crossing detection period in this embodiment, so hereinafter the detection period will also be referred to as the detection period DT.

[0045] The comparator circuit 211A compares the detection period DT with a predetermined value to determine the level, and the de-energization timing setting circuit 212A adjusts the start timing tp of the de-energization period Tp based on the output of the comparator circuit 211A.

[0046] FIG. 5B is a table showing an example of adjusting the non-energized period Tp in the second embodiment.

[0047] In this embodiment, the comparison circuit 211A sets a predetermined value T1 and a predetermined value T2 that is greater than the predetermined value T1 as predetermined values ​​to be compared with the detection period DT. The predetermined value T1 is an example of a first predetermined value, and the predetermined value T2 is an example of a second predetermined value.

[0048] 5B, when the detection period DT is equal to or shorter than the predetermined value T1 (DT≦T1), both the comparison result a1 with the predetermined value T1 and the comparison result a2 with the predetermined value T2 are H level, so the de-energization timing setting circuit 212A determines that the detection period DT is too short and advances the start timing tp of the de-energization period Tp (for example, advances the start timing tp by a change amount ΔT: tp → tp − ΔT). When the detection period DT is longer than the predetermined value T1 but equal to or shorter than the predetermined value T2 (T1<DT≦T2), the comparison result a1 with the predetermined value T1 is L level and the comparison result a2 with the predetermined value T2 is H level, so the de-energization timing setting circuit 212A determines that the detection period DT is appropriate and maintains the start timing tp of the de-energization period Tp (tp → tp). As shown in FIG. 5B, when the detection period DT is longer than the predetermined value T2 (DT > T2), both the comparison result a1 of T1 with the predetermined value and the comparison result a2 of T1 with the predetermined value T2 are at the L level, so the non-energizing timing setting circuit 212A determines that there is too much leeway in the detection period DT and delays the start timing tp of the non-energizing period Tp (for example, delays the start timing tp by a change width ΔT: tp → tp + ΔT).

[0049] The non-energization period Tp includes not only the period required to detect the zero-crossing of the back electromotive force but also a regeneration period during which current flows through the motor windings. Because the regeneration period varies depending on the magnitude of the current flowing through the motor windings, adjusting the start timing tp of the non-energization period Tp using the non-energization period Tp, which includes errors due to such variations, may result in poor accuracy in optimizing the non-energization period Tp. Therefore, the position detection unit 10A generates a detection period signal indicating a detection period DT, which is a portion of the non-energization period Tp and corresponds to the period from the end of the regeneration operation during which current flows through the motor windings to the generation of the position detection signal BEMF. The current waveform control unit (non-energization period designation unit 21A) adjusts the start timing tp of the non-energization period Tp based on the detection period DT indicated by the detection period signal ENB. By adjusting the start timing of the non-energization period Tp using the detection period DT, which excludes the regeneration period from the non-energization period Tp, the non-energization period Tp can be optimized with high accuracy.

[0050] For example, when the current waveform control unit (non-power-carrying period designation unit 21A) detects that the detection period DT is shorter than a predetermined value T1, it advances the start timing tp of the non-power-carrying period Tp. If the detection period DT is too short, it advances the start timing tp of the non-power-carrying period Tp so that the detection period DT becomes longer, thereby optimizing the non-power-carrying period Tp.

[0051] For example, when the current waveform control unit (non-power-on period designation unit 21A) detects that the detection period DT is longer than a predetermined value T2, it delays the start timing tp of the non-power-on period Tp. If the detection period DT is too long, it is possible to optimize the non-power-on period Tp by delaying the start timing tp of the non-power-on period Tp so that the detection period DT becomes shorter.

[0052] The latch circuit 213 latches an H level signal in synchronization with the output of the de-energization timing setting circuit 212A, and is reset by the position detection signal BEMF via an inverter 214. The Q output of the latch circuit 213 becomes the de-energization period signal TP.

[0053] Next, an example of the operation of the motor drive device of this embodiment until the long non-energized period Tp reaches an appropriate value will be described with reference to FIG.

[0054] 6 is a waveform diagram illustrating the operation of the non-energized period designation unit 21A according to the second embodiment. From the top to the bottom, FIG. 6 shows the back electromotive force Bemf, the non-energized period signal TP, the phase potential Vu and the neutral point potential Vc, the detection period signal ENB, and the position detection signal BEMF. The time 0 on the horizontal axis in the diagram indicates the timing at which the position detection signal BEMF rises, and the rising edge of the position detection signal BEMF resets the time information for timing the non-energized period Tp to zero.

[0055] If the start timing tp of the non-energized period Tp is time tp1, the non-energized period Tp starts at time tp1. The detection period DT, which is the period from the end of diode regeneration to the rising edge of the position detection signal BEMF, i.e., the high-level period of the detection period signal ENB, is measured by the timer circuit 210. Assume that the comparison circuit 211A determines that DT > T2. In this case, the non-energized timing setting circuit 212A delays the start timing tp of the next non-energized period Tp by ΔT from time tp1 to time tp2 (tp2 = tp1 + ΔT). If the non-energized period Tp starts at time tp2 of the next cycle and the comparison circuit 211A determines that DT > T2 again in this cycle, the non-energized timing setting circuit 212A delays the start timing tp of the next non-energized period Tp by the change amount ΔT from time tp2, as in the previous cycle. Then, suppose that the de-energization period Tp starts at time tp3 (= tp2 + ΔT) of the next cycle, and the comparator circuit 211A determines that T1 < DT ≦ T2. In this case, the de-energization timing setting circuit 212A maintains the start timing tp of the next de-energization period Tp at time tp3. As a result, the detection period DT is maintained between the predetermined value T1 and the predetermined value T2 thereafter.

[0056] As described above, according to this embodiment, the non-energizing period Tp can be shortened so that the zero-cross detection period DT in sensorless motor drive approaches a target value, thereby enabling low-vibration and low-noise drive.

[0057] In the second embodiment, the predetermined values ​​T1, T2, and the variation width ΔT are also set arbitrarily in design. For example, the variation width ΔT may be different depending on whether the start timing tp is delayed or advanced. Furthermore, the above values ​​do not have to be fixed values ​​and may be changed as appropriate.

[0058] Furthermore, the change in the start timing tp of the non-power-on period Tp does not have to be determined solely by comparing a single detection period DT with a predetermined value, but may be determined by the fact that the comparison results are the same for a predetermined number of consecutive times, or may be determined by the average value of a predetermined number of comparison results.

[0059] For example, if the start timing tp of the non-power-conduction period Tp is advanced each time it is detected that the detection period DT is shorter than the predetermined value T1, the non-power-conduction period Tp will be changed frequently. Therefore, the current waveform control unit (non-power-conduction period designation unit 21A) may advance the start timing tp of the non-power-conduction period Tp when it detects that the detection period DT is shorter than the predetermined value T1 a predetermined number of times. This makes it possible to suppress vibration, noise, or malfunctions that accompany frequent changes to the non-power-conduction period Tp.

[0060] For example, if the start timing tp of the non-power-conduction period Tp is delayed every time it is detected that the detection period DT is longer than the predetermined value T2, the non-power-conduction period Tp will be changed frequently. Therefore, the current waveform control unit (non-power-conduction period designation unit 21A) may delay the start timing tp of the non-power-conduction period Tp when it detects that the detection period DT is longer than the predetermined value T2 a predetermined number of times. This makes it possible to suppress vibration, noise, or malfunctions that accompany frequent changes to the non-power-conduction period Tp.

[0061] Third Embodiment When the detection period DT of the zero crossing of the back electromotive force Bemf is significantly longer than the predetermined value T1, which is the target value, for example, when the detection period DT is several tens to several hundreds of times the variation width ΔT, it may take a long time to shorten the non-power-supply period Tp to an appropriate value if the variation width ΔT is a fixed value. In contrast, in the third embodiment, an example will be described in which the number of determination ranks for the detection period DT is increased and the variation width ΔT is set according to the rank, thereby speeding up the convergence of the non-power-supply period Tp to an appropriate value.

[0062] 7A is a circuit diagram showing an example of a non-energized period designation unit 21B according to the third embodiment. Note that the configuration other than the non-energized period designation unit 21B is the same as that in the second embodiment, and therefore illustration and description thereof will be omitted.

[0063] 5A in that the de-energized period designation unit 21B according to the third embodiment has a comparison circuit 211B and a de-energized timing setting circuit 212B instead of the comparison circuit 211A and the de-energized timing setting circuit 212A. Specifically, in addition to the predetermined values ​​T1 and T2, predetermined values ​​T3 and T4 are provided, the number of ranks of the detection period DT is increased to five, and change widths ΔT, ΔT1, and ΔT2 are provided according to each rank.

[0064] FIG. 7B is a table showing an example of adjusting the non-energized period Tp in the third embodiment.

[0065] In this embodiment, the comparison circuit 211B sets a predetermined value T1, a predetermined value T2 greater than the predetermined value T1, a predetermined value T3 greater than the predetermined value T2, and a predetermined value T4 greater than the predetermined value T3 as predetermined values ​​to be compared with the detection period DT.

[0066] 7B, when the time signal DT is equal to or less than a predetermined value T1 (DT≦T1), the de-energization timing setting circuit 212B determines that the detection period DT is too short and advances the start timing tp of the de-energization period Tp (for example, advances the start timing tp by a change amount ΔT: tp→tp−ΔT). As shown in FIG. 7B, when the time signal DT is longer than the predetermined value T1 and equal to or less than a predetermined value T2 (T1<DT≦T2), the de-energization timing setting circuit 212B determines that the detection period DT is appropriate and maintains the start timing tp of the de-energization period Tp (tp→tp). 7B , when the detection period DT is longer than the predetermined value T2 but not greater than the predetermined value T3 (T2<DT≦T3), the de-energization timing setting circuit 212B determines that there is still time left in the detection period DT and delays the start timing tp of the de-energization period Tp (for example, delaying the start timing tp by a change amount ΔT: tp→tp+ΔT). When the detection period DT is longer than the predetermined value T3 but not greater than the predetermined value T4 (T3<DT≦T4), the de-energization timing setting circuit 212B determines that there is still time left in the detection period DT and increases the change amount by which the start timing tp of the de-energization period Tp is delayed from ΔT to ΔT1, delaying the start timing tp by the change amount ΔT1 (tp→tp+ΔT1). As shown in Figure 7B, if the detection period DT is longer than a predetermined value T4 (DT > T4), the de-energization timing setting circuit 212B further increases the change width to ΔT2 and delays the start timing tp by the change width ΔT2 (tp → tp + ΔT2).

[0067] For example, the predetermined value T2 is set to twice the predetermined value T1 (T2 = 2T1), the predetermined value T3 is set to 12 times the predetermined value T1 (T3 = 12T1), and the predetermined value T4 is set to 92 times the predetermined value T1 (T3 = 92T1), and the change widths are also set to ΔT = T1, ΔT1 = 10T1, and ΔT2 = 90T1. When the detection period DT is equal to or less than the predetermined value T3, it reaches an appropriate value equal to or less than the predetermined value T2 within 10 cycles, and the speed at which the detection period DT reaches the appropriate value is the same as in the second embodiment when ΔT, T1, and T2 are set the same as in this embodiment. When the detection period DT is equal to or less than the predetermined value T4, it reaches the predetermined value T3 within 8 cycles, and this reaching speed (i.e., the change width ΔT1) is 10 times the change width ΔT. When the detection period DT is longer than the predetermined value T4, the speed at which the detection period DT reaches or falls below the predetermined value T4 (i.e., the change width ΔT2) is 90 times the change width ΔT, thereby speeding up the detection period DT reaching the appropriate value.

[0068] Thus, for example, when the current detection period DT is significantly longer than the predetermined value T2, if the amount of change by which the start timing tp of the non-power-conduction period Tp is delayed is small, it will take a long time to optimize the non-power-conduction period Tp. Therefore, the current waveform control unit (non-power-conduction period designation unit 21B) can speed up the optimization of the non-power-conduction period Tp by setting the amount of change by which the start timing tp of the non-power-conduction period Tp is delayed to a larger amount the greater the difference between the detection period DT and the predetermined value T2.

[0069] In this embodiment, an example has been described in which the convergence of the non-power period Tp to an appropriate value is accelerated by increasing the judgment rank of the detection period DT and setting the change width ΔT according to the rank for the motor drive device of the second embodiment, but the convergence of the non-power period Tp to an appropriate value may also be accelerated in a similar manner for the motor drive device of the first embodiment.

[0070] Specifically, in the first embodiment, when the current non-current-carrying period Tp is significantly longer than the predetermined value T2, if the amount of change by which the start timing tp of the non-current-carrying period Tp is delayed is small, it will take a long time to optimize the non-current-carrying period Tp. Therefore, the current waveform control unit 20 sets the amount of change by which the start timing tp of the non-current-carrying period Tp is delayed to a larger amount the greater the difference between the non-current-carrying period Tp and the predetermined value T2, thereby enabling faster optimization of the non-current-carrying period Tp.

[0071] Fourth Embodiment In the second and third embodiments, a predetermined value T1 is set as the target minimum value of the detection period DT, and the start timing tp of the non-energized period Tp is adjusted so that the detection period DT approaches the predetermined value T1. While these methods can adjust the non-energized period Tp, including the detection period DT, to an appropriate value during constant rotation control, for example, during acceleration or deceleration control of the motor 100, a larger current flows through the motor windings than during constant rotation control. Therefore, it is necessary to ensure a sufficiently long non-energized period Tp, including the detection period DT. When the current in the motor windings increases, the regenerative current during the non-energized period Tp also increases, lengthening the regeneration period. If the non-energized period Tp is insufficient, the zero-crossing of the back electromotive force will have already occurred by the time the detection period DT begins after the regeneration operation ends, preventing proper zero-crossing detection. In the fourth embodiment, an example is described in which this problem is prevented.

[0072] 8A is a circuit diagram showing an example of a non-energized period designation unit 21C according to the fourth embodiment. Note that the configuration other than the non-energized period designation unit 21C is the same as that in the second embodiment, and therefore illustration and description thereof will be omitted.

[0073] The de-energization period designation unit 21C according to the fourth embodiment differs from the de-energization period designation unit 21A according to the second embodiment shown in FIG. 5A in that it includes a comparison circuit 211C and a de-energization timing setting circuit 212C instead of the comparison circuit 211A and the de-energization timing setting circuit 212A. Specifically, in addition to the predetermined values ​​T1 and T2, a predetermined value T0 is provided, and if the detection period DT measured by the timer circuit 210 is shorter than the predetermined value T0, the start timing tp of the de-energization period Tp is advanced by a change amount ΔT0 larger than the change amount ΔT up to that point (for example, ΔT0 = 5ΔT). Note that the predetermined value T0 is an example of a third predetermined value.

[0074] FIG. 8B is a table showing an example of adjusting the non-energized period Tp in the fourth embodiment.

[0075] In this embodiment, the comparison circuit 211C sets a predetermined value T1, a predetermined value T2 larger than the predetermined value T1, and a predetermined value T0 smaller than the predetermined value T1 as predetermined values ​​to be compared with the detection period DT.

[0076] An example of the operation of the motor drive device of this embodiment from the time when a zero crossing is erroneously detected until the non-energizing period Tp reaches an appropriate value will be described with reference to FIG.

[0077] 9 is a waveform diagram illustrating the operation of the non-energized period designation unit 21C according to the fourth embodiment. From the top to the bottom, FIG. 9 shows the back electromotive force Bemf, the non-energized period signal TP, the phase potential Vu and the neutral point potential Vc, the detection period signal ENB, and the position detection signal BEMF. Times t11 and t12 on the horizontal axis in the diagram indicate the timings at which the position detection signal BEMF rises, and the rising edge of the position detection signal BEMF resets the time information for timing the non-energized period Tp to zero.

[0078] If the start timing tp of the non-energized period Tp is time tp1, the non-energized period Tp begins at time tp1, and diode regeneration begins, in which the phase potential Vu falls below ground potential. If the diode regeneration ends at time t11, the phase potential Vu rises at time t11. However, at time t11, the zero-crossing point of the back electromotive force Bemf has already passed, and the phase potential Vu is lower than the neutral point potential Vc. As a result, the pulse widths of the detection period signal ENB and the position detection signal BEMF become minimal. Because the detection period DT, whose length corresponds to the minimal pulse width, is shorter than the predetermined value T0, the start timing tp of the next non-energized period Tp, time tp2, is advanced by a large change amount ΔT0 (e.g., 5ΔT).

[0079] Assume that the non-conduction period Tp starts at time tp2, which is advanced by ΔT0, and then, at time t21, the diode regeneration operation ends and the detection period signal ENB rises. At time t21, the back electromotive force Bemf has not yet reached zero, and the phase potential Vu is higher than the neutral point potential Vc. Eventually, at time t22, the back electromotive force Bemf crosses zero, and the position detection signal BEMF rises. Because the start timing tp of the non-conduction period Tp has been significantly advanced, the detection period DT is longer than the predetermined value T2, and the start timing tp of the next non-conduction period Tp, time tp3, is delayed by the change amount ΔT. From this point on, the delay operation of the start timing tp of the non-conduction period Tp in the second embodiment described with reference to FIG. 6 is repeated. If the detection period DT falls between the predetermined values ​​T1 and T2, it is determined to be appropriate and maintained at the appropriate value.

[0080] As described above, when the current waveform control unit (non-energizing period designation unit 21C) detects that the detection period DT is shorter than the predetermined value T0, which is shorter than the predetermined value T1, it sets the amount of change by which the start timing tp of the non-energizing period Tp is advanced to a greater extent than when it detects that the detection period DT is longer than the predetermined value T0. For example, when the detection period DT is short (i.e., when the start timing tp of the non-energizing period Tp is late), the detection period DT may start after the zero-crossing timing of the back electromotive force has passed, and the zero-crossing may not be detected correctly. Therefore, when the detection period DT is shorter than the predetermined value T0, the amount of change by which the start timing tp of the non-energizing period Tp is advanced to a greater extent, making it possible to detect the zero-crossing correctly.

[0081] Specifically, a predetermined value T0 shorter than the target minimum detection period T1 is set, and if the measured detection period DT is shorter than the predetermined value T0, it is determined that the zero-crossing detection is abnormal, and the non-power-on period Tp is temporarily reset to a larger value to redo the optimization of the zero-crossing detection. A state in which the zero-crossing detection is abnormal, i.e., a state in which the position detection is incorrect, can lead to abnormal operation such as loss of synchronization, which is a more serious state than current distortion due to the non-power-on period Tp, and therefore avoiding this state can be given priority.

[0082] Furthermore, when the measured detection period DT is shorter than the predetermined value T0, the predetermined values ​​T1 and T2 may be further extended. An example in which the predetermined values ​​T1 and T2 are extended when the measured detection period DT is shorter than the predetermined value T0 will be described as a modified example of the fourth embodiment with reference to Figures 10A and 10B.

[0083] 10A is a circuit diagram showing an example of a non-energizing period designation unit 21D according to a modification of the fourth embodiment. The non-energizing period designation unit 21D according to the modification of the fourth embodiment differs from the non-energizing period designation unit 21C according to the fourth embodiment shown in FIG. 8A in that a comparison circuit 211D is included instead of the comparison circuit 211C. Specifically, the comparison result a0 is fed back to the units for setting the predetermined values ​​T1 and T2. More specifically, if the detection period DT is shorter than the predetermined value T0, the start timing tp of the non-energizing period Tp is advanced by a change amount ΔT0 (e.g., ΔT0 = 5ΔT) larger than the previous change amount ΔT, and the predetermined values ​​T1 and T2 are each increased by a unit change amount ΔT.

[0084] FIG. 10B is a table showing an example of adjusting the non-energized period Tp in the modified example of the fourth embodiment.

[0085] In this modified example, the comparison circuit 211D sets a predetermined value T1, a predetermined value T2 greater than the predetermined value T1, and a predetermined value T0 smaller than the predetermined value T1 as predetermined values ​​to be compared with the detection period DT, and increases the predetermined values ​​T1 and T2 depending on the situation.

[0086] If the initially set predetermined values ​​T1 and T2 are shorter than the appropriate values, the detection period DT may again become shorter than the predetermined value T0 due to the delay operation of the start timing tp that is repeated after time tp3 of the start timing tp of the non-energized period Tp in Figure 9. Therefore, if the detection period DT becomes shorter than the predetermined value T0, the start timing tp of the next non-energized period Tp is advanced by a change amount ΔT0 = 5ΔT, and the predetermined values ​​T1 and T2 are each further increased by a unit change amount ΔT. By repeating the above operation, the predetermined values ​​T1 and T2 are updated so that the detection period DT falls within the predetermined value, and the detection period DT and the non-energized period settle to the appropriate values.

[0087] Thus, if the predetermined values ​​T1 and T2 are too short, the detection period DT also becomes short, and zero crossings may not be detected correctly. Therefore, if the current waveform control unit (non-energized period designation unit 21D) detects that the detection period DT is shorter than the predetermined value T0, it may reset the predetermined values ​​T1 and T2 to be longer. This allows zero crossings to be detected correctly.

[0088] Fifth Embodiment In the second and subsequent embodiments described above, an example has been described in which the position detection unit 10A includes two comparators: a comparator 11A that compares the phase potential Vu with the neutral point potential Vc, and a comparator 13 that compares the phase potential Vu with the ground potential. However, because the position detection signal BEMF output by the comparator 11A is not generated during the diode regeneration period indicated by the signal CMP1 output by the comparator 13, the comparators 11A and 13 can be realized by switching and operating one comparator.

[0089] FIG. 11 is a circuit diagram showing an example of a position detection unit 10B according to the fifth embodiment.

[0090] 11 , the position detection unit 10B includes a selection circuit 15 that selects and outputs either the neutral point potential Vc or ground potential, a comparator 16 that receives the potential (phase potential Vu) at the junction of the high-side switch 41 and the low-side switch 42 and compares the output of the selection circuit 15 with the phase potential Vu of the U phase, a one-shot pulse circuit 17 that detects the falling edge of the output of the comparator 16 and outputs a one-shot pulse position detection signal BEMF, and an AND circuit 18 that outputs a detection period signal ENB that is the logical product of the output of the comparator 16 and a non-energized period signal TP. The selection circuit 15 selects and outputs the neutral point potential Vc when the output of the comparator 16 is at an H level, and selects and outputs the ground potential when the output of the comparator 16 is at an L level. The comparator 16 is activated by the non-energized period signal TP, and when the non-energized period signal TP is at an L level, the comparator 16 also outputs an L level.

[0091] The operation of the position detector 10B having the above configuration, which receives the non-energized period signal TP, the neutral point potential Vc, and the phase potential Vu, and outputs the position detection signal BEMF and the detection period signal ENB, will be described below.

[0092] When the non-energized period signal TP is at an L level, the comparator 16 also outputs an L level, so the selection circuit 15 selects ground potential. The detection period signal ENB, which is the output of the AND circuit 18, is also at an L level. When the non-energized period signal TP becomes an H level and the switching operation of the U phase of the drive unit 40 stops, the inertia of the motor winding U causes the phase potential Vu to drop sharply below ground potential, causing the body diode of the low-side switch 42 to conduct and perform a regenerative operation in which a regenerative current flows. Because the phase potential Vu is below ground potential by the forward voltage drop of the body diode, the active comparator 16 maintains its output at an L level. Therefore, the detection period signal ENB is also maintained at an L level.

[0093] When the regenerative current flowing through the U phase disappears, the body diode of the low-side switch 42 turns off, and the phase potential Vu rises to become a back electromotive force Bemf superimposed on the neutral point potential Vc, causing the output of the comparator 16 to rise to the H level. When the output of the comparator 16 becomes the H level, the detection period signal ENB also becomes the H level via the AND circuit 18, and the selection circuit 15 selects and outputs the neutral point potential Vc.

[0094] At the zero-crossing point where the polarity of the back electromotive force Bemf becomes negative, the phase potential Vu falls below the neutral point potential Vc, causing the output of the comparator 16 to fall to an L level. The detection period signal ENB also falls to an L level via the AND circuit 18, and a one-shot pulse of the position detection signal BEMF is generated via the one-shot pulse circuit 17. The position detection signal BEMF resets the non-energized period signal TP in the non-energized period designation unit and causes it to fall to an L level. The selection circuit 15 selects the ground potential, but since the non-energized period signal TP falls to an L level, the comparator 16 becomes inactive and outputs an L level. Although not shown, jitter may occur in the output of the comparator 16 depending on the timing of the falling edge of the non-energized period signal TP and the switching of the selection circuit 15. Therefore, a delay may be provided when the selection circuit 15 selects the ground potential.

[0095] In this way, the comparator 16 detects the start timing of the detection period DT by comparing the connection point potential (phase potential Vu) with the ground potential after the start of the non-energized period Tp, and outputs the position detection signal BEMF by comparing the connection point potential (phase potential Vu) with the neutral point potential Vc of the motor 100 after the start of the detection period DT.

[0096] As described above, according to the position detector 10B of the fifth embodiment, a single comparator serves as both a comparator for detecting zero crossings (e.g., comparator 11A) and a comparator for detecting regenerative operation to identify the detection period DT (e.g., comparator 13), thereby simplifying the circuit configuration. The position detectors 10A in the second and subsequent embodiments can be replaced with this position detector 10B.

[0097] (Other Embodiments) As described above, the embodiments have been described as examples of the technology according to the present disclosure. However, the technology according to the present disclosure is not limited to these, and can be applied to embodiments in which modifications, substitutions, additions, omissions, etc. are made as appropriate. For example, the following modifications are also included in one embodiment of the present disclosure.

[0098] For example, in the second and subsequent embodiments, the diode regenerative operation of the low-side switch 42 is detected by comparing the phase potential Vu of the U phase with the ground potential. Thus, during the non-conduction period Tp after the diode regenerative operation of the low-side switch 42, a zero crossing at which the back electromotive force transitions from a positive potential to a negative potential is detected. This configuration allows the comparator circuit to be configured on the ground potential side. On the other hand, although a level shift circuit is required between the low-side switch 42 and other small-signal circuits, the comparator circuit may be configured on the power supply potential side to compare the phase potential Vu with the power supply potential to detect the diode regenerative operation of the high-side switch 41 and detect a zero crossing at which the back electromotive force transitions from a negative potential to a positive potential during the non-conduction period Tp after the diode regenerative operation. In this case, the comparator circuit for comparing the phase potential Vu with the power supply potential generates the detection period signal ENB, using the period when the phase potential Vu is higher than the power supply potential as the detection period DT. In this way, the position detection unit may have a comparison circuit that compares the potential of the junction (phase potential Vu) between the high-side switch 41 and the low-side switch 42 with the power supply potential, and may generate the detection period signal ENB by using the period during the non-energized period Tp in which the potential of the junction (phase potential Vu) is lower than the power supply potential as the detection period DT. For example, in the fifth embodiment, the comparator 16 may detect the start timing of the detection period DT by comparing the potential of the junction (phase potential Vu) with the power supply potential after the start of the non-energized period Tp, and may output the position detection signal BEMF by comparing the potential of the junction (phase potential Vu) with the neutral point potential Vc of the motor 100 after the start of the detection period DT.

[0099] For example, in the above embodiment, the neutral point potential Vc, which is compared with the phase potential Vu to detect the zero crossing of the back electromotive force, is detected directly from the connection point of each motor winding, but the detection method is not limited to this. For example, the neutral point potential Vc may be a virtual neutral point potential obtained by dividing the voltage between the power supply and ground of the drive unit 40.

[0100] For the same reason that the neutral point potential for zero-cross detection may be the potential of a virtual neutral point, zero-cross detection does not necessarily mean detecting the zero point of the back electromotive force strictly. In other words, it is sufficient that zero-cross detection can be performed within the range allowed for motor control.

[0101] For example, in the above embodiment, the non-energizing period Tp or the detection period DT is described as remaining between predetermined values ​​T1 and T2, which are target values. However, the target value is not limited to the range from predetermined value T1 to predetermined value T2 and may be a constant value. In other words, predetermined values ​​T1 and T2 may be the same value (T1 = T2). In this case, the non-energizing period Tp or the detection period DT operates to fluctuate around predetermined value T1. As a result, the length of the non-energizing period Tp changes only within a small range, which effectively suppresses noise or vibration caused by current waveform distortion.

[0102] For example, in the above embodiment, an example was described in which the phase potential Vu was compared with ground potential to detect diode regenerative operation, but the comparison voltage does not need to be exactly ground potential. During diode regenerative operation in the non-energized period Tp, the phase potential Vu is lower than ground potential by the forward voltage of the diode, so the comparison voltage only needs to be between ground potential and ground potential minus the forward voltage of the diode. For the same reason, to detect regenerative operation in which current is drawn from the winding on the high-side switch 41 side, the comparison voltage does not need to be exactly power supply potential, but only needs to be between the power supply potential and the power supply potential plus the forward voltage of the diode.

[0103] In the above-described embodiment, a method for adjusting the start timing tp of the non-energizing period Tp to shorten the non-energizing period Tp was described in which a target value was set for the non-energizing period Tp or the net detection period DT, which is the non-energizing period Tp minus the regeneration period, and the non-energizing period Tp was adjusted to approach the target value. Furthermore, in the fourth embodiment, an example was described in which the target value was updated to optimize the non-energizing period Tp. However, there may be a phase difference between the back electromotive force Bemf and the motor current, and in this case, there is a limit to how much the regeneration period can be shortened.

[0104] Therefore, by correcting the current phase by using an advance angle, the non-power-conduction period Tp can be further shortened. For example, the time from the start of the non-power-conduction period Tp to the end of regeneration (the regeneration period) and the time from the end of regeneration to the zero crossing (the detection period DT) are measured. If there is sufficient time in the detection period DT, the start timing tp of the next non-power-conduction period Tp is delayed by ΔT. Alternatively, if the detection period DT is insufficient, the start timing tp of the next non-power-conduction period Tp is advanced by ΔT, and the current is advanced by an amount equivalent to the regeneration period. Because the advance angle shortens the detection period DT and the regeneration period, the start timing tp of the non-power-conduction period Tp approaches the zero-crossing detection point. Ideally, the regeneration period is almost eliminated, and the non-power-conduction period Tp can be brought closer to the target value of the detection period DT. This will be described using Figures 12A and 12B.

[0105] 12A and 12B are waveform diagrams illustrating the operation of the non-energization period designation unit according to another embodiment. FIG. 12A shows the waveform before a current advance angle is added, and FIG. 12B shows the waveform after a current advance angle is added. In FIG. 12A , the detection period DT is between predetermined values ​​T1 and T2, achieving the optimized state described in the previous embodiments. However, if there is a phase difference between the current Iu and the back electromotive force Bemf, the current Iu is large at the start timing t1 of the non-energization period Tp, resulting in a long regeneration period. Therefore, the non-energization period Tp is also measured. If it is determined that the non-energization period Tp can be shortened when the detection period DT falls between predetermined values ​​T1 and T2 (specifically, if the non-energization period Tp is equal to or greater than a fourth predetermined value longer than the predetermined value T2), the current Iu is advanced by the regeneration period (Tp - DT), as shown in FIG. 12B . For example, the fourth predetermined value is equal to or greater than twice the predetermined value T2, and when Tp≧2×T2, the current Iu is advanced by the regeneration period (Tp−DT). By advancing the current, the current Iu at the start of the non-energization period Tp becomes smaller, shortening the subsequent regeneration period, but lengthening the detection period DT until the zero-crossing point. Therefore, the start timing tp of the non-energization period Tp is delayed again so that the extended detection period DT becomes an appropriate value. When the detection period DT falls within the target range, a determination is made as to whether to further shorten or maintain the regeneration period. By repeating the above process, both the regeneration period and the detection period DT can be shortened, further shortening the non-energization period Tp, which is the sum of the two.

[0106] In this manner, the current waveform control unit may control the current waveform to advance the phase by a predetermined value when the detection period DT is shorter than the predetermined value T2 and the non-energization period Tp is equal to or greater than a fourth predetermined value longer than the predetermined value T2. When the detection period DT is shorter than the predetermined value T2 and the non-energization period Tp is equal to or greater than the fourth predetermined value, a regeneration period occurs, and the current flowing through the motor windings is large at the start timing tp of the non-energization period Tp. Therefore, by advancing the current phase, the current flowing through the motor windings can be reduced at the start timing tp of the non-energization period Tp, thereby shortening the regeneration period. This further shortens the non-energization period Tp, further suppressing vibration, noise, and malfunction.

[0107] In the above description, the lead angle value of the current Iu is the regeneration period (Tp-DT), but this is not limiting. For example, the non-energized period Tp may be used as the lead angle value.

[0108] For example, in the fourth embodiment, the current waveform control unit (non-energizing period designation unit 21C) sets the amount of change by which the start timing tp is advanced to a greater extent when it detects that the detection period DT is shorter than the predetermined value T0, which is shorter than the predetermined value T1, than when it detects that the detection period DT is longer than the predetermined value T0. For example, when the current waveform control unit detects that the detection period DT is longer than a fifth predetermined value, which is longer than the predetermined value T2, the current waveform control unit may set the amount of change by which the start timing of the non-energizing period Tp is advanced to a greater extent than when it detects that the detection period DT is shorter than the fifth predetermined value. Furthermore, when the current waveform control unit detects that the detection period DT is longer than the fifth predetermined value, the current waveform control unit may reset the predetermined values ​​T1 and T2 to be shorter.

[0109] For example, in the above embodiment, the control for optimizing the non-energized period Tp has been described with a focus on the motor winding U, but similar control for optimizing the non-energized period Tp is also performed for the motor windings V and W. Note that when the rotation speed of the motor 100 is stable, the control of the present disclosure for optimizing the non-energized period Tp may be performed for one of the motor windings U, V, and W, and the same control as for the one motor winding may be performed for the other motor windings.

[0110] In the above-described embodiment, each component included in the motor drive device may be configured with dedicated hardware, or may be realized by executing a software program suitable for each component. Each component may be realized by a program execution unit such as a CPU or processor reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory.

[0111] Some or all of the functions of the motor drive device according to the above embodiments are typically realized as an LSI, which is an integrated circuit. These may be individually integrated into single chips, or some or all of them may be integrated into a single chip. Furthermore, the integrated circuit is not limited to an LSI, and may be realized using a dedicated circuit or a general-purpose processor. It is also possible to use an FPGA (Field Programmable Gate Array), which can be programmed after LSI manufacturing, or a reconfigurable processor, which allows the connections and settings of circuit cells within the LSI to be reconfigured.

[0112] Furthermore, if an integrated circuit technology that can replace LSI emerges due to advances in semiconductor technology or other derivative technologies, it is natural that each component included in the motor drive device can be integrated using that technology.

[0113] In addition, this disclosure also includes forms obtained by making various modifications to the embodiments that a person skilled in the art would think of, and forms realized by arbitrarily combining the components and functions in each embodiment within the scope of the present disclosure.

[0114] (Additional Notes) The above description of the embodiments discloses the following techniques.

[0115] (Technology 1) A motor drive device for driving a sensorless motor, comprising: a position detection unit that detects back electromotive force generated in a motor winding during a non-energized period when no current is applied to the motor winding, and outputs a position detection signal; a current waveform control unit that controls a current waveform including the non-energized period; an energization control unit that controls energization to the motor winding based on the current waveform; and a drive unit that supplies current to the motor winding in accordance with control of the energization control unit, wherein the energization control unit switches the energized phase of a motor that includes the motor winding based on the position detection signal, and the current waveform control unit adjusts the start timing of the non-energized period based on the non-energized period.

[0116] According to this, if the non-energized period, which is the period from when the power supply to the motor windings is stopped until the zero crossing of the back electromotive force is detected, is too long, the start timing of the non-energized period can be adjusted to shorten the non-energized period. If the non-energized period is too short, the start timing of the non-energized period can be adjusted to lengthen the non-energized period. For example, instead of selecting the start timing of the non-energized period from predetermined candidates, the start timing of the non-energized period can be flexibly adjusted by feedback control of the non-energized period. Therefore, the non-energized period can be optimized. For example, the non-energized period can be shortened while being set to the period necessary for rotor position detection, thereby suppressing vibration, noise, and malfunction.

[0117] (Technology 2) The motor drive device according to Technology 1, wherein the current waveform control unit advances the start timing when it detects that the non-energized period is shorter than a first predetermined value.

[0118] According to this, if the non-energizing period is too short, the start timing of the non-energizing period can be advanced so that the non-energizing period is extended, thereby optimizing the non-energizing period.

[0119] (Technology 3) The motor drive device according to Technology 2, wherein the current waveform control unit advances the start timing when it detects that the non-energized period is shorter than the first predetermined value a predetermined number of times.

[0120] For example, if the start timing of the non-power-conduction period is advanced each time it is detected that the non-power-conduction period is shorter than the first predetermined value, the non-power-conduction period will be changed frequently. Therefore, by advancing the start timing of the non-power-conduction period when it is detected a predetermined number of times that the non-power-conduction period is shorter than the first predetermined value, it is possible to suppress vibrations, noise, or malfunctions that are associated with frequent changes in the non-power-conduction period.

[0121] (Technology 4) The motor drive device according to Technology 1, wherein the current waveform control unit delays the start timing when it detects that the non-energized period is longer than a second predetermined value.

[0122] According to this, if the non-energized period is too long, the start timing of the non-energized period is delayed so as to shorten the non-energized period, thereby making it possible to optimize the non-energized period.

[0123] (Technology 5) The motor drive device according to Technology 4, wherein the current waveform control unit delays the start timing when it detects that the non-energized period is longer than the second predetermined value a predetermined number of times.

[0124] For example, if the start timing of the non-energized period is delayed each time it is detected that the non-energized period is longer than the second predetermined value, the non-energized period will be changed frequently. Therefore, by delaying the start timing of the non-energized period when it is detected a predetermined number of times that the non-energized period is longer than the second predetermined value, it is possible to suppress vibrations, noise, or malfunctions that are associated with frequent changes in the non-energized period.

[0125] (Technology 6) The motor drive device according to Technology 4 or 5, wherein the current waveform control unit sets a larger amount of change for delaying the start timing as the difference between the non-energized period and the second predetermined value increases.

[0126] For example, if the current non-energizing period is significantly longer than the second predetermined value, and the amount of change by which the start timing of the non-energizing period is delayed is small, it will take a long time to optimize the non-energizing period. Therefore, by setting the amount of change by which the start timing of the non-energizing period is delayed to a larger amount the greater the difference between the non-energizing period and the second predetermined value, it is possible to speed up the optimization of the non-energizing period.

[0127] (Technology 7) A motor drive device according to Technology 1, wherein the drive unit has a configuration in which a high-side switch and a low-side switch, each having a body diode or a diode connected in anti-parallel, are connected in series between a power supply and ground, the drive unit alternately turns on the high-side switch and the low-side switch at a predetermined frequency during periods other than the non-energized period, and turns off both the high-side switch and the low-side switch during the non-energized period, the position detection unit generates a detection period signal indicating a detection period, which is a part of the non-energized period and is a period from the end of a regenerative operation in which current flows through the motor windings to the generation of the position detection signal, and the current waveform control unit adjusts the start timing based on the detection period indicated by the detection period signal.

[0128] The non-energized period includes not only the period required to detect the zero-crossing of the back electromotive force but also the regeneration period, during which current flows through the motor windings. Because the regeneration period varies depending on the magnitude of the current flowing through the motor windings, adjusting the start timing of the non-energized period using a non-energized period that includes errors due to such variations may result in poor accuracy in optimizing the non-energized period. Therefore, by adjusting the start timing of the non-energized period using a detection period that excludes the regeneration period from the non-energized period, it is possible to optimize the non-energized period with high accuracy.

[0129] (Technology 8) The motor drive device according to Technology 7, wherein the current waveform control unit advances the start timing when it detects that the detection period is shorter than a first predetermined value.

[0130] According to this, if the detection period is too short, the start timing of the non-energized period can be advanced so that the detection period becomes longer, thereby optimizing the non-energized period.

[0131] (Technology 9) The motor drive device according to Technology 8, wherein the current waveform control unit advances the start timing when it detects that the detection period is shorter than the first predetermined value a predetermined number of times.

[0132] For example, if the start timing of the non-energized period is advanced each time it is detected that the detection period is shorter than the first predetermined value, the non-energized period will be changed frequently. Therefore, by advancing the start timing of the non-energized period when it is detected that the detection period is shorter than the first predetermined value a predetermined number of times, it is possible to suppress vibration, noise, or malfunction that would be associated with frequent changes in the non-energized period.

[0133] (Technology 10) A motor drive device according to Technology 8 or 9, wherein when the current waveform control unit detects that the detection period is shorter than a third predetermined value that is shorter than the first predetermined value, the amount of change by which the start timing is advanced is set to be greater than when the current waveform control unit detects that the detection period is longer than the third predetermined value.

[0134] For example, if the detection period is short (i.e., if the start timing of the non-energized period is late), the detection period may start after the zero-cross timing of the back electromotive force has passed, and the zero-cross may not be detected correctly. Therefore, if the detection period is shorter than the third predetermined value, the amount of change that advances the start timing of the non-energized period is set to a large value, so that the zero-cross can be detected correctly.

[0135] (Technology 11) A motor drive device according to Technology 10, wherein the current waveform control unit resets the first predetermined value and a second predetermined value greater than the first predetermined value to be longer when it detects that the detection period is shorter than the third predetermined value.

[0136] If the first and second predetermined values ​​are too short, the detection period will also be short, and zero crossings may not be detected correctly. Therefore, if the detection period is shorter than the third predetermined value, the first and second predetermined values ​​can be reset to be longer, thereby enabling zero crossings to be detected correctly.

[0137] (Technology 12) The motor drive device according to Technology 7, wherein the current waveform control unit delays the start timing when it detects that the detection period is longer than a second predetermined value.

[0138] According to this, if the detection period is too long, the start timing of the non-energized period is delayed so as to shorten the detection period, thereby making it possible to optimize the non-energized period.

[0139] (Technology 13) The motor drive device according to Technology 12, wherein the current waveform control unit delays the start timing when it detects that the detection period is longer than the second predetermined value a predetermined number of times.

[0140] For example, if the start timing of the non-energized period is delayed each time it is detected that the detection period is longer than the second predetermined value, the non-energized period will be changed frequently. Therefore, by delaying the start timing of the non-energized period when it is detected that the detection period is longer than the second predetermined value a predetermined number of times, it is possible to suppress vibrations, noise, or malfunctions that are associated with frequent changes to the non-energized period.

[0141] (Technology 14) A motor drive device according to Technology 12 or 13, wherein the current waveform control unit sets a larger amount of change for delaying the start timing as the difference between the detection period and the second predetermined value increases.

[0142] For example, if the current detection period is significantly longer than the second predetermined value, and the amount of change by which the start timing of the non-energized period is delayed is small, it will take a long time to optimize the non-energized period. Therefore, by setting the amount of change by which the start timing of the non-energized period is delayed to a larger amount as the difference between the detection period and the second predetermined value increases, it is possible to speed up the optimization of the non-energized period.

[0143] (Technology 15) A motor drive device according to any one of Technologies 7 to 14, wherein the position detection unit has a comparison circuit that compares the potential of the connection point between the high-side switch and the low-side switch with ground potential, and generates the detection period signal by defining a period during the non-energized period when the potential of the connection point is higher than the ground potential as the detection period.

[0144] In this way, by comparing the potential at the connection point with the ground potential, the detection period can be measured.

[0145] (Technology 16) A motor drive device according to any one of Technologies 7 to 14, wherein the position detection unit has a comparison circuit that compares the potential of the connection point between the high-side switch and the low-side switch with the power supply potential, and generates the detection period signal by using the period during the non-energized period when the potential of the connection point is lower than the power supply potential as the detection period.

[0146] In this way, by comparing the potential at the connection point with the power supply potential, the detection period can be measured.

[0147] (Technology 17) A motor drive device according to any one of Technologies 7 to 14, wherein the position detection unit has a comparator to which a potential at a connection point between the high-side switch and the low-side switch is input, the comparator detects the start timing of the detection period by comparing the potential at the connection point with ground potential or power supply potential after the start of the non-energized period, and outputs the position detection signal by comparing the potential at the connection point with a neutral point potential of the motor after the start of the detection period.

[0148] In this way, by using a single comparator to both detect zero crossings and detect regenerative operation to identify the detection period, the circuit configuration can be simplified.

[0149] (Technology 18) A motor drive device described in any one of Techniques 7 to 17, wherein the current waveform control unit controls the current waveform to advance the phase by a predetermined value when the detection period is shorter than a second predetermined value and the non-current-carrying period is equal to or greater than a fourth predetermined value that is longer than the second predetermined value.

[0150] According to this, if the detection period is shorter than the second predetermined value and the non-energized period is equal to or greater than the fourth predetermined value, a regeneration period is occurring, and the current flowing through the motor windings is large at the start of the non-energized period. Therefore, by advancing the current phase, the current flowing through the motor windings can be reduced at the start of the non-energized period, thereby shortening the regeneration period. This further shortens the non-energized period, thereby further suppressing vibration, noise, and malfunction.

[0151] The motor drive device according to the present disclosure can be applied to spindle motors of various disk devices that are required to drive sensorless motors with low vibration and low noise.

[0152] 10, 10A, 10B Position detection unit 11, 11A, 13, 16 Comparator 12, 17 One-shot pulse circuit 14, 18 AND circuit 15 Selection circuit 20 Current waveform control unit 21, 21A, 21B, 21C, 21D De-energization period designation unit 210 Time counting circuit 211, 211A, 211B, 211C, 211D Comparison circuit 212, 212A, 212B, 212C De-energization timing setting circuit 213 Latch circuit 214 Inverter 22 Waveform generation unit 30 Energization control unit 40 Drive unit 41 High-side switch 42 Low-side switch 100 Motor

Claims

1. A motor drive device for driving a sensorless motor, comprising: a position detection unit that outputs a position detection signal by detecting back electromotive force generated in the motor windings during a non-energization period when no current is applied to the motor windings; a current waveform control unit that controls a current waveform including the non-energization period; an energization control unit that controls energization to the motor windings based on the current waveform; and a drive unit that supplies current to the motor windings under control of the energization control unit, wherein the energization control unit switches the energization phase of a motor that has the motor windings based on the position detection signal, and the current waveform control unit adjusts the start timing of the non-energization period based on the non-energization period.

2. The motor drive device according to claim 1, wherein the current waveform control section advances the start timing when it detects that the non-energizing period is shorter than a first predetermined value.

3. The motor drive device according to claim 2, wherein the current waveform control section advances the start timing when it detects that the non-energizing period is shorter than the first predetermined value a predetermined number of times.

4. The motor drive device according to claim 1, wherein the current waveform control section delays the start timing when it detects that the non-energizing period is longer than a second predetermined value.

5. The motor drive device according to claim 4, wherein the current waveform control section delays the start timing when it detects that the non-energizing period is longer than the second predetermined value a predetermined number of times.

6. The motor drive device according to claim 4 or 5, wherein the current waveform control section sets a larger amount of change for delaying the start timing as the difference between the non-energized period and the second predetermined value increases.

7. The motor drive device according to claim 1, wherein the drive section has a configuration in which a high-side switch and a low-side switch, each having a body diode or an anti-parallel connected diode, are connected in series between a power supply and ground, and alternately turns on the high-side switch and the low-side switch at a predetermined frequency during periods other than the non-energized period and turns off both the high-side switch and the low-side switch during the non-energized period; the position detection section generates a detection period signal indicating a detection period, which is a part of the non-energized period and is the period from the end of regenerative operation in which current flows through the motor winding to the generation of the position detection signal; and the current waveform control section adjusts the start timing based on the detection period indicated by the detection period signal.

8. The motor drive device according to claim 7, wherein the current waveform control section advances the start timing when it detects that the detection period is shorter than a first predetermined value.

9. The motor drive device according to claim 8, wherein the current waveform control section advances the start timing when it detects that the detection period is shorter than the first predetermined value a predetermined number of times.

10. A motor drive device according to claim 8 or 9, wherein when the current waveform control unit detects that the detection period is shorter than a third predetermined value that is shorter than the first predetermined value, the current waveform control unit sets the amount of change by which the start timing is advanced to a value greater than when the current waveform control unit detects that the detection period is longer than the third predetermined value.

11. The motor drive device according to claim 10, wherein the current waveform control unit resets the first predetermined value and the second predetermined value, which is greater than the first predetermined value, to be longer when it detects that the detection period is shorter than the third predetermined value.

12. The motor drive device according to claim 7, wherein the current waveform control section delays the start timing when it detects that the detection period is longer than a second predetermined value.

13. The motor drive device according to claim 12, wherein the current waveform control section delays the start timing when it detects that the detection period is longer than the second predetermined value a predetermined number of times.

14. The motor drive device according to claim 12 or 13, wherein the current waveform control section sets a larger amount of change for delaying the start timing as the difference between the detection period and the second predetermined value increases.

15. The motor drive device according to any one of claims 7 to 14, wherein the position detection unit has a comparison circuit that compares the potential at the connection point between the high-side switch and the low-side switch with ground potential, and generates the detection period signal by defining the period during the non-energized period in which the potential at the connection point is higher than ground potential as the detection period.

16. A motor drive device according to any one of claims 7 to 14, wherein the position detection unit has a comparison circuit that compares the potential at the connection point between the high-side switch and the low-side switch with the power supply potential, and generates the detection period signal by defining the period during the non-energized period when the potential at the connection point is lower than the power supply potential as the detection period.

17. A motor drive device according to any one of claims 7 to 14, wherein the position detection unit has a comparator to which a potential at a connection point between the high-side switch and the low-side switch is input, and the comparator detects the start timing of the detection period by comparing the potential at the connection point with ground potential or power supply potential after the start of the non-energized period, and outputs the position detection signal by comparing the potential at the connection point with a neutral point potential of the motor after the start of the detection period.

18. A motor drive device according to any one of claims 7 to 17, wherein the current waveform control unit controls the current waveform so as to advance the phase by a predetermined value when the detection period is shorter than a second predetermined value and the non-energized period is equal to or greater than a fourth predetermined value that is longer than the second predetermined value.

Images