JP4243567B2 - Sensorless motor driving apparatus and driving method thereof - Google Patents

Description

  The present invention relates to a sensorless motor driving apparatus and a driving method thereof, and more particularly, to a sensorless motor start-up control by separately excited commutation control.

In a brushless motor, electrical commutation is performed instead of mechanical commutation by a brush. Electrical commutation requires the position of the rotor (ie, the angle of rotation).
A conventional brushless motor is provided with a position sensor such as a Hall element, and the position of the rotor is detected through the position sensor (see, for example, Patent Documents 1 and 2).

In the sensorless motor, a voltage induced in the motor coil (hereinafter referred to as BEMF (back electromotive force)) is detected during rotation of the rotor. The position of the rotor is detected without using the position sensor by using the BEMF.
Since a sensorless motor does not have a position sensor, the number of parts is small and wiring is relatively simple. Therefore, since it is easy to reduce the size, it is frequently used as a spindle motor such as an FDD, HDD, and MD / CD / DVD drive, and as a cylinder motor such as a VTR and a video camera.

FIG. 27 is a block diagram showing a conventional sensorless motor driving device (see, for example, Non-Patent Document 1). This sensorless motor driving apparatus drives, for example, a three-phase sensorless motor M.
The PWM control unit 10 generates a PWM control signal P and a PWM mask signal MPWM based on the position signal PS. Here, the position signal PS indicates the estimated position of the rotor (not shown) of the sensorless motor M. The PWM control signal P indicates the energization timing of the motor coils Mu, Mv, Mw of the sensorless motor M. The PWM mask signal MPWM is maintained active for a predetermined time from each of the rising edge and falling edge of the PWM control signal P.

The pre-drive circuit 20 selects, for example, one of the three high-side power transistors 31U, 31V, 31W of the output circuit 3 according to the energization phase switching signal CP and maintains it in the on state. Further, one of the three low-side power transistors 32U, 32V, and 32W is selected and turned on and off in small increments according to the PWM control signal P. As a result, two of the three motor coils Mu, Mv, and Mw are energized. Here, there are six types of energization patterns, each corresponding to a different energized phase.
A magnetic field is generated in the energized motor coil, and torque is applied to the rotor.

The BEMF comparison unit 4 includes three comparators 4U, 4V, and 4W. The potential of each of the three drive terminals U0, V0, and W0 of the sensorless motor M and the potential of the midpoint C of the three motor coils Mu, Mv, and Mw (hereinafter referred to as “below”) , Referred to as midpoint voltage).
The self-excited commutation circuit 50 detects the coincidence between the potentials of the drive terminals U0, V0, and W0 and the midpoint voltage of the motor coil based on the three output signals BCU, BCV, and BCW of the BEMF comparison unit 4. Here, the output signals BCU, BCV, BCW of the BEMF comparator 4 are masked according to two types of mask signals MZC and MPWM. Thereby, in each of the energized phases, the coincidence (hereinafter referred to as zero cross) of the BEMF in the non-energized motor coil and the midpoint voltage of the motor coil is accurately detected. The self-excited commutation circuit 50 generates a self-excited commutation signal SC when the zero cross is detected.

The separately excited commutation circuit 60 generates a constant pulse signal, that is, a separately excited commutation signal FC at a predetermined period.
Count unit 70 selects either self-excited commutation signal SC or other-excited commutation signal FC, and sends the selected commutation signal CS to energized phase switching circuit 80. On the other hand, the interval of the commutation signal CS is measured, and the position signal PS is generated based on the interval. Here, the commutation signal CS is set, for example, so that the phase is delayed by about 30 ° from the selected self-excited commutation signal SC or the separately-excited commutation signal FC.
The energized phase switching circuit 80 generates an energized phase switching signal CP for each input of the commutation signal CS. Further, the position detection mask signal MZC is kept active for a predetermined time from the input time of the commutation signal CS.

  FIG. 28 shows three motor coils Mu, Mv, and Mw current (ie, phase currents) Iu, Iv, Iw, and BEMFVu, Vv, Vw, and three drives for the sensorless motor M when the rotor rotates stably. FIG. 5 is a waveform diagram showing potentials VU0, VV0, VW0 and position detection mask signal MZC of terminals U0, V0, W0, respectively. In FIG. 28, the horizontal axis indicates the phase and corresponds to the position of the rotor expressed in electrical angle.

  The predrive circuit 20 switches on and off the six power transistors of the output circuit 3 for each energized phase. Here, there are six types of on-off patterns and accompanying motor coil energization patterns from I to VI shown in FIG. That is, the energized phase divides one cycle of the phase current into 60 ° increments. In each energized phase, one of the three motor coils Mu, Mv, and Mw has a source current (current flowing in the direction from the drive terminal to the middle point C of the motor coil (the direction of the arrow shown in FIG. 27)). The sink current (current flowing in the direction opposite to the source current) flows to the other one, and the remaining one is kept non-energized. In FIG. 28, the sink current is indicated by hatching.

  In the conventional sensorless motor driving device, the pre-drive circuit 20 is in each energized phase, for example, the high-side power transistors 31U, 31V, 31W through which the source current flows are kept on, and the low-side power transistors 32U, 32V through which the sink current flows. PWM control for on / off of 32W. Further, the pre-drive circuit 20 performs hard switching to turn on and off the power transistor in accordance with switching of the energized phase. Thereby, the phase currents Iu, Iv, and Iw form a rectangular wave. In particular, in each of the motor coils Mu, Mv, and Mw, a 120 ° energization period and a 60 ° non-energization period are alternately repeated. Further, the phase difference between the phase currents Iu, Iv, and Iw is maintained at 120 °.

By the rotation of the rotor, BEMFVu, Vv, and Vw are induced in the motor coils Mu, Mv, and Mw, respectively. The BEMFVu, Vv, and Vw waveforms are close to sine waves.
The potentials VU0, VV0, VW0 of the three drive terminals U0, V0, W0 of the sensorless motor M are equal to voltages obtained by superimposing BEMFVu, Vv, Vw on the drive voltage applied from the output circuit 3, respectively. Here, in FIG. 28, detailed ripples of the output voltages VU0, VV0, and VW0 by PWM control are omitted.
In each energized phase, BEMF Vu, Vv, and Vw cause a zero cross in a non-energized motor coil, that is, a motor coil to which no drive voltage is applied from the output circuit 3 (see white circle ZC shown in FIG. 28). Therefore, when the potentials VU0, VV0, VW0 of the drive terminals U0, V0, W0 coincide with the midpoint voltage of the motor coil (see the black circle ZC0 shown in FIG. 28), the zero crossings ZC of BEMFVu, Vv, Vw respectively Correspond.

The levels of BEMFVu, Vv, and Vw are determined by the magnetic pole center of the stator (part where the magnetic fields generated by the phase currents Iu, Iv, and Iw flowing through the motor coils Mu, Mv, and Mw are particularly concentrated) and the magnetic pole of the rotor. Corresponds to the angle between the center. In particular, at the time of zero crossing, the position of the rotor coincides with one of six predetermined positions (0 °, 60 °, 120 °, 180 °, 240 °, 300 ° in FIG. 28) at 60 ° (electrical angle) intervals. .
Therefore, the position of the rotor is estimated by detecting the zero cross ZC0 by the potentials VU0, VV0, VW0 of the drive terminals U0, V0, W0, respectively. The energization of the motor coil is controlled based on the estimated rotor position, and for example, the angle between the magnetic pole center of the stator and the magnetic pole center of the rotor is maintained within an appropriate range. As a result, the torque is efficiently applied to the rotor and maintained sufficiently large.
In FIG. 28, since the phase of the commutation signal CS is delayed by about 30 ° from the time point of detection of the zero cross, the energized phase is switched with a time delay of about 30 ° from the time of detection of the zero cross. Thereby, the angle between the magnetic pole center of the stator and the magnetic pole center of the rotor is maintained substantially equal to 90 ° (electrical angle), so that the torque applied to the rotor is maintained at the maximum. Therefore, stable rotation of the rotor is efficiently maintained and strong against load fluctuations.

Changes in the potentials VUO, VV0, and VW0 of the drive terminals U0, V0, and W0 actually include noise due to the on / off of the power transistor. The noise is mainly noise N (see FIG. 28) accompanying switching of the energized phase and fine ripple (not shown) accompanying PWM control. Those noises give an error to the detection of the zero cross by the BEMF comparison unit 4.
In the conventional sensorless motor driving apparatus, the position detection mask signal MZC is kept active for a predetermined time every time each energized phase is switched, that is, every time the commutation signal CS is generated (see FIG. 28). During that period, the output signals BCU, BCV, BCW of the BEMF comparison unit 4 are masked, so that it is possible to avoid erroneous detection of zero cross due to noise N accompanying switching of energized phases.
Further, every time the PWM control signal P rises / falls, the PWM mask signal MPWM is kept active for a predetermined time. During that period, the output signals BCU, BCV, BCW of the BEMF comparison unit 4 are masked, so that erroneous detection of zero cross due to ripple accompanying PWM control is avoided.
Thus, accurate detection of zero crossing is realized.

As described above, the position detection of the rotor in the sensorless motor is based on the detection of the zero cross by BEMF in the motor coil. Since BEMF is detected only when the rotor rotates at a certain speed or higher, the above-described rotor position detection cannot be used to start a sensorless motor.
The conventional sensorless motor driving device uses, for example, the separately excited commutation signal FC from the separately excited commutation circuit 60 instead of the self excited commutation signal SC from the self excited commutation circuit 50 when the sensorless motor is started (see FIG. 27). Thereby, the energization of the motor coil is switched at a constant cycle regardless of the actual position of the rotor. The energization control based on the separately excited commutation signal FC (hereinafter referred to as separately excited commutation control) is continued from the start time of starting the sensorless motor for a predetermined time or until the rotation speed of the rotor reaches a constant value. Thereafter, the separately excited commutation control is switched to energization control (hereinafter referred to as self-excited commutation control) based on the self-excited commutation signal SC.
Thus, the conventional sensorless motor driving apparatus realizes the start of the sensorless motor without detecting the position of the rotor.

JP 2003-174789 JP 2003-244983 A Satoshi Kusaka, “Special Feature: Design of Motor Control Circuits Learned through Experiments, Chapter 4: Driving Method of Sensorless DC Motor”, Transistor Technology, CQ Publisher, February 2000, Vol. 37, No. 425, p.221-228

Generally, in a brushless motor, as the energized phase is switched, the attractive force / repulsive force between the magnetic pole of the stator and the magnetic pole of the rotor, and the respective stress distributions in the stator and the rotor change periodically. As a result, noise (hereinafter referred to as motor echo noise) is generally generated from the brushless motor. In particular, as shown in FIG. 28, when the phase currents Iu, Iv, Iw show steep changes, the motor echo noise tends to become excessive. Therefore, a gradual change in phase current is desirable for suppressing motor echo noise.
Unlike a sensorless motor, a brushless motor having a position sensor does not require a motor coil non-energization period to detect the position of the rotor. Accordingly, since a gradual change in phase current can be easily realized, it is relatively easy to suppress motor echo noise (see, for example, Patent Documents 1 and 2). In particular, when the energization control of the motor coil is performed by PWM control, the detection accuracy of the position sensor is maintained high regardless of the ripple accompanying the PWM control.

On the other hand, in the sensorless motor, as described above, the zero cross detection of BEMF requires a non-energization period of the motor coil. However, the gradual change in phase current shortens the non-energization period of the motor coil. In particular, in energization control similar to a brushless motor having a position sensor, it is not possible to ensure a non-energization period of the motor coil.
Furthermore, ripples associated with PWM control reduce the zero-cross detection accuracy and must be reliably masked. However, a gradual change in phase current requires extension of the PWM control period. Along with this, the period required for the mask is extended, so that the zero-cross detection period is shortened.
Thus, in sensorless motors, suppression of motor echo noise hindered accurate detection of zero crossing.

  The hindrance to the accurate detection of zero crossing prevented the improvement of the reliability of the self-excited commutation control, and in particular the increase in the torque generated by the sensorless motor. In other words, it is difficult to achieve both suppression of motor echo noise and increase in torque generated by self-excited commutation control.

The hindrance to accurate detection of zero crossing further made quick and reliable start-up of sensorless motors difficult as follows.
In the separately excited commutation control, the energization of the motor coil is switched regardless of the actual position of the rotor. Accordingly, since the angle between the magnetic pole center of the stator and the magnetic pole center of the rotor is generally out of the optimum range, it is generally difficult to increase the generated torque. As a result, in starting a sensorless motor by separately excited commutation control, it is difficult to increase the starting torque, so it is difficult to shorten the starting time. Furthermore, the start-up control was weak against load fluctuation.
For example, when a sensorless motor is used as a spindle motor of a CD / DVD combined drive, the moment of inertia differs between CD and DVD. Further, when a sensorless motor is used as the HDD spindle motor, the number of magnetic disks varies depending on the capacity, and the disk radius varies depending on the size. In separately excited commutation control, it is difficult to stabilize the start of the sensorless motor in common for any of such various loads.
In order to eliminate these difficulties, it is necessary to switch from the separately excited commutation control to the self-excited commutation control as quickly and reliably as possible when starting the sensorless motor.

However, in the conventional sensorless motor driving device, the separately excited commutation control is continued when the sensorless motor is activated, for example, for a predetermined time from the activation time or until the rotation speed of the rotor reaches a constant value. That is, the separately-excited commutation control cannot be switched to the self-excited commutation control until a state where accurate detection of zero crossing is considered possible. When accurate detection of zero crossing is hindered, shortening the predetermined time is difficult because it impairs the certainty of activation. On the other hand, the time required from the time when the sensorless motor is started to the time when the rotational speed of the rotor reaches a certain value is difficult to shorten by separately excited commutation control.
Thus, in the conventional sensorless motor driving device, it is difficult to quickly and surely switch from the separately excited commutation control to the self-excited commutation control when starting the sensorless motor. Therefore, it is difficult to start the sensorless motor quickly and reliably.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a sensorless motor driving apparatus and a driving method thereof that accurately detect BEMF zero crossing while suppressing motor echo noise, thereby realizing quick and reliable start of the sensorless motor.

The sensorless motor driving apparatus according to the first aspect of the present invention is:
An output circuit for energizing the motor coil of the sensorless motor;
A PWM control unit that generates a PWM control signal indicating the timing of energization of the motor coil based on a position signal indicating the estimated position of the rotor of the sensorless motor;
Energized phase switching circuit that switches the energized phase in synchronization with the commutation signal;
A pre-drive circuit that selects a motor coil corresponding to the energized phase, changes the energization of the selected motor coil by the output circuit according to the PWM control signal, and prohibits the energization of a specific motor coil by the output circuit, particularly in the PWM inhibition period ;
A BEMF comparator that detects BEMF induced in the motor coil and compares the BEMF with the midpoint voltage of the motor coil;
A self-excited commutation circuit that detects coincidence of the BEMF and the midpoint voltage of the motor coil during the BEMF detection period, that is, detects a zero cross, and generates a self-excited commutation signal when the zero cross is detected;
The commutation signal is generated based on the self-excited commutation signal, the interval of the commutation signal is measured, the position signal is generated based on the interval, and the PWM inhibition period and the BEMF detection period are based on the position signal. In particular, a counting unit that starts the PWM prohibition period earlier than the BEMF detection period and ends both the PWM prohibition period and the BEMF detection period in synchronization with the commutation signal;
Have

This sensorless motor driving apparatus first sets a PWM prohibition period, and prohibits energization of a specific motor coil during that period. Thus, a non-energization period of the motor coil is ensured.
Next, the BEMF detection period is started after the start of the PWM prohibition period. Thereby, at the start of the BEMF detection period for one of the motor coils, the PWM control for energization of the motor coil is already stopped. Therefore, accurate detection of zero crossing is possible.
Further, in synchronization with the commutation signal, that is, when the zero cross is detected, the energized phase is switched, and the PWM prohibition period and the BEMF detection period are ended. As a result, when the zero cross is detected, the energization of the motor coil and the PWM control for the motor coil can be resumed quickly. For example, when energization of the motor coil is controlled so that the phase current changes slowly, the start-up of the phase current can be started simultaneously with the detection of zero crossing. Accordingly, since the phase current is sufficiently large during the period in which the rotor is located within a range suitable for torque generation, the generated torque is sufficiently large.
Thus, the sensorless motor driving apparatus according to the first aspect of the present invention can suppress the motor echo noise and sufficiently increase the generated torque due to the gradual change of the phase current.

In the sensorless motor driving apparatus according to the first aspect of the present invention, preferably the PWM control unit,
A command circuit for setting a target current based on the original command and the position signal;
A current comparator that detects a current of the motor coil, that is, a phase current, and compares the detected current with a target current; and a PWM control circuit that generates a PWM control signal according to a difference between the detected current and the target current;
Have That is, the PWM control unit performs feedback control regarding the phase current.
Particularly preferably, the command circuit gradually increases or decreases the target current. Thereby, the phase current is controlled to a gentle waveform, so that motor echo noise is suppressed.

In the sensorless motor driving device according to the first aspect of the present invention, preferably, when a certain time has elapsed from the start of the PWM inhibition period, or the estimated position of the rotor changes by a certain amount from the value at the start of the PWM inhibition period. The counting unit starts the BEMF detection period at the earlier time point.
As a result, when the rotation of the rotor is low, the BEMF detection period is started when a fixed time has elapsed from the start of the PWM prohibition period. On the other hand, when the rotor rotates at a high speed, the BEMF detection period starts when the estimated position of the rotor changes by a certain amount from the value at the start of the PWM inhibition period. As a result, the BEMF detection period is ensured sufficiently long regardless of the actual number of rotations of the rotor, so that zero cross can be accurately detected.

The sensorless motor driving apparatus according to the second aspect of the present invention is
An output circuit for energizing the motor coil of the sensorless motor;
A PWM control unit that generates a PWM control signal indicating the timing of energization of the motor coil based on a position signal indicating the estimated position of the rotor of the sensorless motor;
Energized phase switching circuit that switches the energized phase in synchronization with the commutation signal;
A pre-drive circuit that selects a motor coil corresponding to an energized phase and changes the energization of the selected motor coil by an output circuit in accordance with a PWM control signal;
A BEMF comparator that detects BEMF induced in the motor coil and compares the BEMF with a midpoint voltage of the motor coil;
A self-commutation circuit that detects coincidence of the BEMF and the midpoint voltage during the BEMF detection period, that is, detects a zero cross and generates a self-excited commutation signal when the zero cross is detected;
A separately excited commutation circuit for generating a separately excited commutation signal at a predetermined period;
A selection circuit that selects, as a commutation signal, one of the self-commutation signal and the other excitation commutation signal that is input first during the BEMF detection period; and
A counting circuit that measures the interval of the commutation signal, generates the position signal based on the interval, sets the BEMF detection period based on the position signal, and ends the BEMF detection period in synchronization with the commutation signal;
Have

This sensorless motor drive device secures a BEMF detection period and performs a zero-cross detection operation in parallel during the separately excited commutation control. The commutation signal is a signal generated earlier of the self-excited commutation signal or the separately-excited commutation signal, and the BEMF detection period ends with the generation of the commutation signal. Since the self-excited commutation signal is generated only during the BEMF detection period, the self-excited commutation signal is always selected as the commutation signal when the self-excited commutation signal is generated, that is, when a zero cross is detected. Thus, the separately excited commutation control is promptly switched to the self-excited commutation control by detecting the zero cross.
In particular, when the sensorless motor is started by the separately excited commutation control, switching from the separately excited commutation control to the self-excited commutation control is performed quickly and reliably. Therefore, the sensorless motor can be started quickly and reliably.

In the sensorless motor driving apparatus according to the second aspect of the present invention, preferably the PWM control unit,
A command circuit for setting a target current based on the original command and the position signal;
A current comparator that detects a current of the motor coil, that is, a phase current, and compares the detected current with a target current; and a PWM control circuit that generates a PWM control signal according to a difference between the detected current and the target current;
Have That is, the PWM control unit performs feedback control regarding the phase current.
Particularly preferably, the command circuit gradually increases or decreases the target current. Thereby, the phase current is controlled to a gentle waveform, so that motor echo noise is suppressed.

In the sensorless motor driving device according to the second aspect of the present invention, preferably,
A count circuit sets a PWM prohibition period based on the position signal, in particular, starts the PWM prohibition period earlier than the BEMF detection period, and ends the PWM prohibition period in synchronization with the commutation signal;
The pre-drive circuit prohibits the energization of a specific motor coil by the output circuit during the PWM prohibition period.

This sensorless motor driving apparatus first sets a PWM prohibition period, and prohibits energization of a specific motor coil during that period. Thus, a non-energization period of the motor coil is ensured.
Next, the BEMF detection period is started after the start of the PWM prohibition period. Thereby, at the start of the BEMF detection period for one of the motor coils, the PWM control for energization of the motor coil is already stopped. Therefore, accurate detection of zero crossing is possible.
Further, in synchronization with the commutation signal, that is, when the zero cross is detected, the energized phase is switched, and the PWM prohibition period and the BEMF detection period are ended. As a result, when the zero cross is detected, the energization of the motor coil and the PWM control for the motor coil can be resumed quickly. For example, when energization of the motor coil is controlled so that the phase current changes slowly, the start-up of the phase current can be started simultaneously with the detection of zero crossing. Accordingly, since the phase current is sufficiently large during the period in which the rotor is located within a range suitable for torque generation, the generated torque is sufficiently large.
Thus, the sensorless motor driving apparatus according to the second aspect of the present invention can suppress the motor echo noise and sufficiently increase the generated torque due to the gradual change of the phase current. In particular, when the sensorless motor is started, the starting torque can be sufficiently increased, so that the starting time can be easily shortened, and the starting control is strong against load fluctuations.

In the sensorless motor driving apparatus according to the second aspect of the present invention, preferably, when a certain time has elapsed from the start of the PWM inhibition period, or the estimated position of the rotor changes by a certain amount from the value at the start of the PWM inhibition period. The counting unit starts the BEMF detection period at the earlier time point.
As a result, when the rotation of the rotor is low, the BEMF detection period is started when a fixed time has elapsed from the start of the PWM prohibition period. On the other hand, when the rotor rotates at a high speed, the BEMF detection period starts when the estimated position of the rotor changes by a certain amount from the value at the start of the PWM inhibition period. As a result, the BEMF detection period is ensured sufficiently long regardless of the actual number of rotations of the rotor, so that zero cross can be accurately detected.
In particular, when the sensorless motor is started, switching from the separately excited commutation control to the self-excited commutation control is performed quickly and reliably, so that the sensorless motor can be started quickly and surely.

In the sensorless motor driving apparatus according to the second aspect of the present invention, preferably, the separately excited commutation circuit constantly generates the separately excited commutation signal.
For example, when a sudden external vibration / shock prevents stable rotation of the rotor, thereby suddenly interrupting the generation of the self-excited commutation signal, the separately excited commutation signal is immediately selected as the commutation signal. . That is, self-excited commutation control is immediately switched to separately-excited commutation control. Therefore, the sensorless motor can be restarted smoothly.

In the sensorless motor driving apparatus according to the second aspect of the present invention, preferably, when the self-excited commutation signal is input prior to the other excitation commutation signal during the BEMF detection period, the other excitation commutation is performed in the next BEMF detection period. The circuit extends the period of the separately excited commutation signal.
Once the zero cross is detected and the self-excited commutation signal is generated, it is highly likely that the zero cross is detected and the self-excited commutation signal is generated even in the next BEMF detection period. Therefore, by extending the period of the separately-excited commutation signal as described above, the possibility of detecting a zero cross in the next BEMF detection period is further enhanced. As a result, the separately excited commutation control can be quickly and reliably switched to the self-excited commutation control with only one zero cross detection. In this way, the sensorless motor can be started quickly and reliably.

In the sensorless motor driving apparatus according to the second aspect of the present invention, preferably, when the separately excited commutation signal is input before the self-excited commutation signal for a predetermined number of times during the BEMF detection period, the next BEMF detection is performed. In the period, the separately excited commutation circuit extends the period of the separately excited commutation signal. Particularly preferably, the entire predetermined number of energized phases corresponds to a period of 360 ° in electrical angle.
When the selection of the separately excited commutation signal continues as described above, the separately excited commutation signal is not synchronized with the rotation of the rotor. In particular, when the sensorless motor is started, it is highly likely that the cycle of the separately excited commutation signal is extremely shorter than the actual rotation cycle of the rotor. For example, such a phenomenon is likely to occur when the load is large. Therefore, by extending the period of the separately excited commutation signal as described above, the period of the separately excited commutation signal is brought closer to the actual rotation period of the rotor. Thereby, the detection possibility of zero crossing increases. As a result, switching from separately excited commutation control to self-excited commutation control can be performed quickly and reliably. In this way, the sensorless motor can be started quickly and reliably.

In the sensorless motor driving apparatus according to the second aspect of the present invention, preferably, the separately excited commutation circuit generates at least two types of pulse signals having different periods, and any one of these pulse signals is separately excited. Select as. Further preferably, the pulse signals include a first signal having a constant period and a second signal having a period twice as long as the period of the first signal.
For example, when the sensorless motor is activated by the separately excited commutation control based on the first signal, if the self-excited commutation signal is not generated easily, the first signal is not synchronized with the rotation of the rotor. In particular, it is highly possible that the period of the first signal is extremely shorter than the actual rotation period of the rotor. For example, such a phenomenon is likely to occur when the load is large. At that time, the first signal is switched to the second signal. Since it is highly possible that the period of the second signal is close to the actual rotation period of the rotor, the possibility of detecting zero crossing is increased. As a result, switching from separately excited commutation control to self-excited commutation control can be performed quickly and reliably. Therefore, the sensorless motor can be started quickly and reliably.

The sensorless motor driving method according to the first aspect of the present invention is:
Generating a PWM control signal indicating the energization timing of the motor coil of the sensorless motor based on the position signal indicating the estimated position of the rotor of the sensorless motor;
Selecting a motor coil corresponding to the energized phase and energizing the selected motor coil in accordance with a PWM control signal;
Starting a PWM inhibition period based on the position signal;
Prohibiting energization of a specific motor coil during the PWM inhibition period;
Detecting BEMF induced in the motor coil and comparing the BEMF with the midpoint voltage of the motor coil;
Starting the BEMF detection period later than the start of the PWM inhibition period;
Detecting the coincidence of the BEMF and the midpoint voltage of the motor coil during the BEMF detection period, that is, zero crossing, and generating a self-excited commutation signal when the zero cross is detected;
Generating a commutation signal based on the self-excited commutation signal;
Ending both the PWM inhibition period and the BEMF detection period in synchronization with the commutation signal;
Switching the energized phase in synchronization with the commutation signal;
Measuring the interval of the commutation signal; and
Generating a position signal based on the spacing of the commutation signals;
Have

In this sensorless motor driving method, first, a PWM prohibition period is set, and the energization of a specific motor coil is prohibited during that period. Thus, a non-energization period of the motor coil is ensured.
Next, the BEMF detection period is started after the start of the PWM prohibition period. Thereby, at the start of the BEMF detection period for one of the motor coils, the PWM control for energization of the motor coil is already stopped. Therefore, accurate detection of zero crossing is possible.
Further, in synchronization with the commutation signal, that is, when the zero cross is detected, the energized phase is switched, and the PWM prohibition period and the BEMF detection period are ended. As a result, when the zero cross is detected, the energization of the motor coil and the PWM control for the motor coil can be resumed quickly. For example, when energization of the motor coil is controlled so that the phase current changes slowly, the start-up of the phase current can be started simultaneously with the detection of zero crossing. Accordingly, since the phase current is sufficiently large during the period in which the rotor is located within a range suitable for torque generation, the generated torque is sufficiently large.
Thus, the sensorless motor driving method according to the first aspect of the present invention can suppress the motor echo noise and increase the generated torque sufficiently by the gradual change of the phase current.

In the sensorless motor driving method according to the first aspect of the present invention, preferably, the step of generating the PWM control signal includes:
A sub-step for setting the target current based on the original command and the position signal;
Detecting a current of the motor coil and comparing the detected current with a target current; and
A sub-step of generating a PWM control signal according to the difference between the detected current and the target current;
including. That is, the PWM control corresponds to feedback control related to the phase current.
Particularly preferably, the target current gradually increases or decreases for each sub-step for setting the target current. Thereby, the phase current is controlled to a gentle waveform, so that motor echo noise is suppressed.

The sensorless motor driving method according to the first aspect of the present invention preferably includes a step of measuring an elapsed time and a change amount of the estimated position of the rotor from the start time of the PWM inhibition period. At that time, BEMF is detected at the earlier of a certain time after the start of the PWM prohibition period, or when the estimated position of the rotor changes by a certain amount from the value at the start of the PWM prohibition period. The step of starting the period is executed.
As a result, when the rotation of the rotor is low, the BEMF detection period is started when a fixed time has elapsed from the start of the PWM prohibition period. On the other hand, when the rotor rotates at a high speed, the BEMF detection period starts when the estimated position of the rotor changes by a certain amount from the value at the start of the PWM inhibition period. As a result, the BEMF detection period is ensured sufficiently long regardless of the actual number of rotations of the rotor, so that zero cross can be accurately detected.

The sensorless motor driving method according to the second aspect of the present invention is as follows.
Generating a PWM control signal indicating the energization timing of the motor coil of the sensorless motor based on the position signal indicating the estimated position of the rotor of the sensorless motor;
Selecting a motor coil corresponding to the energized phase and energizing the selected motor coil in accordance with a PWM control signal;
Detecting BEMF induced in the motor coil and comparing the BEMF with the midpoint voltage of the motor coil;
Starting a BEMF detection period based on the position signal;
Detecting the coincidence of the BEMF and the midpoint voltage of the motor coil during the BEMF detection period, that is, zero crossing, and generating a self-excited commutation signal when the zero cross is detected;
Generating a separately excited commutation signal at a predetermined period;
A step of selecting a self-commutated commutation signal or a separately-excited commutation signal that is generated earlier during the BEMF detection period as a commutation signal;
Ending the BEMF detection period in synchronization with the commutation signal;
Switching the energized phase in synchronization with the commutation signal;
Measuring the interval of the commutation signal; and
Generating a position signal based on the spacing of the commutation signals;
Have

This sensorless motor driving method ensures a BEMF detection period and performs a zero-cross detection operation in parallel during separately excited commutation control. The commutation signal is a signal generated earlier of the self-excited commutation signal or the separately-excited commutation signal, and the BEMF detection period ends with the generation of the commutation signal. Since the self-excited commutation signal is generated only during the BEMF detection period, the self-excited commutation signal is always selected as the commutation signal when the self-excited commutation signal is generated, that is, when a zero cross is detected. Thus, the separately excited commutation control is promptly switched to the self-excited commutation control by detecting the zero cross.
In particular, when the sensorless motor is started by the separately excited commutation control, switching from the separately excited commutation control to the self-excited commutation control is performed quickly and reliably. Therefore, the sensorless motor can be started quickly and reliably.

In the sensorless motor driving method according to the second aspect of the present invention, preferably, the step of generating the PWM control signal includes:
A sub-step for setting the target current based on the original command and the position signal;
Detecting a current of the motor coil and comparing the detected current with a target current; and
A sub-step of generating a PWM control signal according to the difference between the detected current and the target current;
including. That is, the PWM control corresponds to feedback control related to the phase current.
Particularly preferably, the target current gradually increases or decreases for each sub-step for setting the target current. Thereby, the phase current is controlled to a gentle waveform, so that motor echo noise is suppressed.

The sensorless motor driving method according to the second aspect of the present invention is preferably
Starting the PWM inhibition period earlier than the BEMF detection period based on the position signal;
Prohibiting energization of a specific motor coil during the PWM inhibition period; and
Ending the PWM inhibition period in synchronization with the commutation signal;
Have

In this sensorless motor driving method, first, a PWM prohibition period is set, and the energization of a specific motor coil is prohibited during that period. Thus, a non-energization period of the motor coil is ensured.
Next, the BEMF detection period is started after the start of the PWM prohibition period. Thereby, at the start of the BEMF detection period for one of the motor coils, the PWM control for energization of the motor coil is already stopped, so that the zero cross can be accurately detected.
Further, in synchronization with the commutation signal, that is, when the zero cross is detected, the energized phase is switched, and the PWM prohibition period and the BEMF detection period are ended. As a result, when the zero cross is detected, the energization of the motor coil and the PWM control for the motor coil can be resumed quickly. For example, when energization of the motor coil is controlled so that the phase current changes slowly, the start-up of the phase current can be started simultaneously with the detection of zero crossing. Accordingly, since the phase current is sufficiently large during the period in which the rotor is located within a range suitable for torque generation, the generated torque is sufficiently large.
Thus, the sensorless motor driving method according to the second aspect of the present invention can suppress the motor echo noise and increase the generated torque sufficiently by the gradual change of the phase current. In particular, when the sensorless motor is started, the starting torque can be sufficiently increased, so that the starting time can be easily shortened, and the starting control is strong against load fluctuations.

The sensorless motor driving method according to the second aspect of the present invention preferably includes a step of measuring an elapsed time and a change amount of the estimated position of the rotor from the start time of the PWM inhibition period. At that time, BEMF is detected at the earlier of a certain time after the start of the PWM prohibition period, or when the estimated position of the rotor changes by a certain amount from the value at the start of the PWM prohibition period. The step of starting the period is executed.
As a result, when the rotation of the rotor is low, the BEMF detection period is started when a fixed time has elapsed from the start of the PWM prohibition period. On the other hand, when the rotor rotates at a high speed, the BEMF detection period starts when the estimated position of the rotor changes by a certain amount from the value at the start of the PWM inhibition period. As a result, the BEMF detection period is ensured sufficiently long regardless of the actual number of rotations of the rotor, so that zero cross can be accurately detected.
In particular, when the sensorless motor is started, switching from the separately excited commutation control to the self-excited commutation control is performed quickly and reliably, so that the sensorless motor can be started quickly and surely.

In the sensorless motor driving method according to the second aspect of the present invention, preferably, the step of generating the separately excited commutation signal is constantly executed.
For example, when a sudden external vibration / shock prevents stable rotation of the rotor, thereby suddenly interrupting the generation of the self-excited commutation signal, the separately excited commutation signal is immediately selected as the commutation signal. . That is, self-excited commutation control is immediately switched to separately-excited commutation control. Therefore, the sensorless motor can be restarted smoothly.

In the sensorless motor driving method according to the second aspect of the present invention, preferably, when the self-excited commutation signal is generated prior to the other excitation commutation signal during the BEMF detection period, the other excitation commutation is performed in the next BEMF detection period. Extending the period of the signal.
Once the zero cross is detected and the self-excited commutation signal is generated, it is highly likely that the zero cross is detected and the self-excited commutation signal is generated even in the next BEMF detection period. Therefore, by extending the period of the separately-excited commutation signal as described above, the possibility of detecting a zero cross in the next BEMF detection period is further enhanced. As a result, the separately excited commutation control can be quickly and reliably switched to the self-excited commutation control with only one zero cross detection. In this way, the sensorless motor can be started quickly and reliably.

In the sensorless motor driving method according to the second aspect of the present invention, preferably, when the separately excited commutation signal is generated before the self-excited commutation signal for a predetermined number of times during the BEMF detection period, A step of extending the period of the separately excited commutation signal in the detection period. Particularly preferably, the entire predetermined number of energized phases corresponds to a period of 360 ° in electrical angle.
When the selection of the separately excited commutation signal continues as described above, the separately excited commutation signal is not synchronized with the rotation of the rotor. In particular, when the sensorless motor is started, it is highly likely that the cycle of the separately excited commutation signal is extremely shorter than the actual rotation cycle of the rotor. For example, such a phenomenon is likely to occur when the load is large. Therefore, by extending the period of the separately excited commutation signal as described above, the period of the separately excited commutation signal is brought closer to the actual rotation period of the rotor. Thereby, the detection possibility of zero crossing increases. As a result, switching from separately excited commutation control to self-excited commutation control can be performed quickly and reliably. In this way, the sensorless motor can be started quickly and reliably.

In the sensorless motor driving apparatus according to the second aspect of the present invention, preferably, the step of generating the separately excited commutation signal includes a sub-step of generating at least two types of pulse signals having different periods, and any one of those pulse signals. Selecting one as a separately excited commutation signal. Further preferably, the pulse signals include a first signal having a constant period and a second signal having a period twice as long as the period of the first signal.
For example, when the sensorless motor is activated by the separately excited commutation control based on the first signal, if the self-excited commutation signal is not generated easily, the first signal is not synchronized with the rotation of the rotor. In particular, it is highly possible that the period of the first signal is extremely shorter than the actual rotation period of the rotor. For example, such a phenomenon is likely to occur when the load is large. At that time, the first signal is switched to the second signal. Since it is highly possible that the period of the second signal is close to the actual rotation period of the rotor, the possibility of detecting zero crossing is increased. As a result, switching from separately excited commutation control to self-excited commutation control can be performed quickly and reliably. Therefore, the sensorless motor can be started quickly and reliably.

The sensorless motor driving apparatus and method according to the first aspect of the present invention starts the PWM inhibition period earlier than the BEMF detection period. Further, when the zero cross is detected, the energized phase is switched and the PWM inhibition period and the BEMF detection period are ended. Therefore, for example, when motor echo noise is suppressed by a gradual change in phase current, a non-energization period of the motor coil is secured, so that zero cross can be accurately detected. Furthermore, since the phase current quickly rises simultaneously with the detection of the zero cross, the generated torque can be sufficiently increased. Thereby, in particular, the drive control of the sensorless motor is strong against load fluctuation.
Therefore, the sensorless motor driving apparatus and method according to the first aspect of the present invention are used, for example, as a spindle motor for a CD / DVD combined drive and as a general-purpose spindle motor for HDDs having various capacities / sizes. This is advantageous for driving a sensorless motor.

The sensorless motor driving apparatus and method according to the second aspect of the present invention ensures the BEMF detection period and performs the zero cross detection operation in parallel during the separately excited commutation control. Furthermore, by selecting the signal generated earlier of the self-excited commutation signal or the self-excited commutation signal as the commutation signal, it is possible to quickly change from the self-excited commutation control to the self-excited commutation control at the time of zero cross detection. Realize reliable switching. Therefore, the sensorless motor can be started quickly and reliably, particularly when the sensorless motor is started by the separately excited commutation control.
Therefore, the sensorless motor driving apparatus and method according to the second aspect of the present invention is advantageous in improving the reproduction quality when driving a sensorless motor used as a spindle motor of a CD / DVD drive, for example. Furthermore, driving a sensorless motor used as a HDD spindle motor is advantageous in improving the processing speed.

DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, exemplary embodiments of the invention will be described with reference to the drawings.
Embodiment 1
FIG. 1 is a block diagram showing a sensorless motor driving apparatus according to Embodiment 1 of the present invention. This sensorless motor drive device targets, for example, a three-phase (U phase, V phase, W phase) sensorless motor M as a driving target. The sensorless motor M has, for example, three motor coils Mu, Mv, Mw of Y connection, three drive terminals U0, V0, W0, and a midpoint terminal C of the motor coil.
A sensorless motor driving apparatus according to Embodiment 1 of the present invention includes a PWM control unit 1, a pre-drive circuit 2, an output circuit 3, a BEMF comparison unit 4, a self-excited commutation circuit 5, a separately-excited commutation circuit 6, a count unit 7, And an energized phase switching circuit 8.

In the output circuit 3, three sets of two power transistors connected in series are connected in parallel between a power supply terminal 33 maintained at a constant high potential and a ground terminal. The six power transistors are preferably MOSFETs. In addition, an IGBT or a bipolar transistor may be used. A freewheel diode is connected to each of the power transistors. The freewheeling diode is preferably the body diode of the corresponding power transistor. In addition, a diode independent of the power transistor may be used.
Connection points between the high-side power transistors 31U, 31V, 31W and the low-side power transistors 32U, 32V, 32W are connected to three drive terminals U0, V0, W0 of the sensorless motor M, respectively. Further, a current detection resistor R is connected between the low-side power transistors 32U, 32V, 32W and the ground terminal.

The PWM control unit 1 includes an oscillation circuit 11, a torque command circuit 12, a current comparison unit 13, and a PWM control circuit 14.
The oscillation circuit 11 generates two set pulse signals SP1 and SP2. The two set pulse signals SP1 and SP2 preferably have the same frequency (PWM control carrier frequency). Furthermore, the phase difference between the two is maintained at a constant value, preferably 180 °.

The torque command circuit 12 receives the original torque command TQ from, for example, an external microprocessor. The original torque command TQ is preferably an analog signal, and its level indicates the target value of the torque generated by the sensorless motor M, that is, the target value of the output current I from the output circuit 3 to the sensorless motor M.
The torque command circuit 12 further generates an increase torque command TQ1 and a decrease torque command TQ2 based on the original torque command TQ as follows. Here, the increase torque command TQ1 and the decrease torque command TQ2 are preferably analog signals similar to the original torque command TQ, and their levels are two currents in the same direction among the three phase currents Iu, Iv, Iw. Indicates the target value. At that time, the level of the original torque command TQ indicates the target value of the remaining one phase current. The actual correspondence between the original torque command TQ, the increase torque command TQ1, and the decrease torque command TQ2 and the target values of the three phase currents Iu, Iv, Iw is determined for each energized phase by the pre-drive circuit 2 ( See below for details.)

FIG. 2 is a block diagram showing an internal configuration of the torque command circuit 12. As shown in FIG. The torque command circuit 12 includes a torque map circuit 121 and a torque synthesis circuit 122.
The torque map circuit 121 stores a predetermined data map. The data map shows each level change pattern of the increase torque command TQ1 and the decrease torque command TQ2. The pattern is preferably represented by an increasing sequence of ratios to the level of the original torque command TQ for the level of the increasing torque command TQ1, and a decreasing sequence of ratios to the level of the original torque command TQ for the level of the decreasing torque command TQ2. Is done. Here, the sum of the level of the increase torque command TQ1 and the level of the decrease torque command TQ2 is equal to the level of the original torque command TQ.
The torque map circuit 121 further receives the original torque command TQ, and determines the levels of the increase torque command TQ1 and the decrease torque command TQ2 based on the pattern indicated by the data map and the level of the original torque command TQ. Preferably, the level of the original torque command TQ is multiplied by a series of ratios read from the data map, and a series of multiplication values obtained thereby is set as a series of levels of the increase torque command TQ1 and the decrease torque command TQ2. .

The torque synthesis circuit 122 actually generates an increase torque command TQ1 and a decrease torque command TQ2. In particular, each level is changed in synchronization with the position signal PS according to a pattern determined by the torque map circuit 121.
FIG. 3 is a waveform diagram of the position signal PS, the original torque command TQ, the increase torque command TQ1, and the decrease torque command TQ2.
The position signal PS is preferably 16 parallel signals. The number of signal lines may be other numbers, for example, 8 or 32. The position signal PS includes 16 rectangular pulse waves. In FIG. 3, numbers 0, 1, 2,..., 15 are assigned to these pulse waves. These pulse waves have a pulse interval substantially equal to 60 ° (electrical angle) and a pulse width substantially equal to 60 ° / 16 = 3.75 ° (electrical angle). Furthermore, the phase is delayed by 3.75 ° (electrical angle) in the order of numbers 0 to 15.
The leading pulse wave (number 0) is synchronized with the switching of the energized phase (see below for details). Therefore, a series of pulses of numbers 0 to 15 divides the period of one energized phase into 16 sections by 3.75 ° (electrical angle). That is, each of the pulses of numbers 0 to 15 indicates the estimated position of the rotor of the sensorless motor M at an interval of 3.75 ° (electrical angle).

The data map of the torque map circuit 121 preferably indicates an increasing sequence comprising 10 ratios of 1/11, 2/11, 3/11,..., 10/11 for the increasing torque command TQ1, and the decreasing torque command For TQ2, a decreasing sequence consisting of 10 ratios of 10/11, 9/11, 8/11,. At that time, the torque map circuit 121 sets a series of levels of the increase torque command TQ1 to 1/11 times, 2/11 times,..., 10/11 times the level of the original torque command TQ, and a series of reduction torque commands TQ2. Is set to 10/11 times, 9/11 times,..., 1/11 times the level of the original torque command TQ.
First, the torque synthesizing circuit 122 sequentially associates each of the 10 levels of the increase torque command TQ1 and the decrease torque command TQ2 with the pulse waves numbered 0 to 9 of the position signal PS. Next, the original torque command TQ is cut out by a pulse width of 3.75 ° (electrical angle) of the position signal PS in synchronization with the rising of each of the pulse waves of the position signals PS of numbers 0 to 9. Furthermore, the level of the cut out original torque command TQ is converted into a level corresponding to each of the pulse waves of numbers 0 to 9, and sequentially sent to the current comparison unit 13 (see FIG. 1) as an increase torque command TQ1 and a decrease torque command TQ2. To do.
Thus, the level of the increase torque command TQ1 is changed from 1/11 times to 11/11 times the level of the original torque command TQ (see the one-dot chain line shown in FIG. 3) in the period TP of 37.5 ° (electrical angle) from the switching point of the energized phase. 11 times = Increase by 1/11 times the original torque command TQ level to the same magnification (see the solid line shown in Fig. 3). On the other hand, the level of the decrease torque command TQ2 decreases in the same period TP from 10/11 times the level of the original torque command TQ to 1/11 times the original torque command TQ (see the broken line shown in FIG. 3).
Further, outside the period TP described above, the level of the increase torque command TQ1 is maintained at the level of the original torque command TQ, and the level of the decrease torque command TQ2 is maintained at 0.

  The length of the increase / decrease period TP between the increase torque command TQ1 and the decrease torque command TQ2, the increase in the level of the increase torque command TQ1, and the decrease in the level of the decrease torque command TQ2 are set to values different from the above. May be. In particular, the increase / decrease amount of the level may change within the above period TP. These changes can be easily realized by changing the data map in the torque map circuit 121. Further, the levels of the increase torque command TQ1 and the decrease torque command TQ2 may be increased or decreased continuously. However, in any of the above cases, the sum of the level of the increase torque command TQ1 and the level of the decrease torque command TQ2 is maintained equal to the level of the original torque command TQ.

  The current comparison unit 13 includes three comparators 131, 132, and 133. The three comparators 131, 132, and 133 respectively compare the levels of the increase torque command TQ1, the decrease torque command TQ2, and the original torque command TQ with the voltage drop amount at the current detection resistor R. Since the output current I of the output circuit 3 flows through the current detection resistor R, the amount of voltage drop at the current detection resistor R corresponds to the level of the output current I. On the other hand, the levels of the increase torque command TQ1, the decrease torque command TQ2, and the original torque command TQ indicate target values of the three phase currents Iu, Iv, and Iw. Accordingly, the three output signals IC1, IC2, and IC3 of the current comparison unit 13 indicate differences between the target values of the three phase currents Iu, Iv, and Iw and the output current I of the output circuit 3, respectively.

  First, the PWM control circuit 14 raises the first PWM control signal P1 in synchronization with the rise of the first set pulse signal SP1. Thereafter, when the first output signal IC1 of the current comparison unit 13 indicates a level match between the increase torque command TQ1 and the voltage drop amount at the current detection resistor R, the PWM control circuit 14 outputs the first PWM control signal P1. Fall down. Next, the second PWM control signal P2 is raised in synchronization with the rise of the second set pulse signal SP2. After that, when the second output signal IC2 of the current comparison unit 13 indicates a level match between the decrease torque command TQ2 and the voltage drop amount at the current detection resistor R, the PWM control circuit 14 outputs the second PWM control signal P2. Fall down.

When one of the first PWM control signal P1 and the second PWM control signal P2 falls before the other falls, the PWM control circuit 14 determines whether the first output signal IC1 of the current comparison unit 13 and the second output signal IC1 Are masked together with the output signal IC2. After that, when the third output signal IC3 of the current comparison unit 13 indicates the coincidence of the level between the original torque command TQ and the voltage drop amount at the current detection resistor R, the PWM control circuit 14 only detects the second PWM control signal P2. Fall down. Subsequently, the PWM control circuit 14 cancels the masking for the first output signal IC1 and the second output signal IC2.
Here, the PWM control circuit 14 may lower only the first PWM control signal P1 instead of the second PWM control signal P2. In addition, the PWM control signal to be lowered may be determined for each period of a predetermined electrical angle (for example, 30 °).

The PWM control circuit 14 keeps the PWM mask signal MPWM active for a certain period from the rising edge and falling edge of the first PWM control signal P1.
The two PWM control signals P1 and P2 are sent to the pre-drive circuit 2, and the PWM mask signal MPWM is sent to the self-excited commutation circuit 5.

The predrive circuit 2 controls on / off of the six power transistors of the output circuit 3 for each energized phase according to the energized phase switching signal CP as follows, and controls energization of the motor coils Mu, Mv, and Mw.
Here, there are six types of energization patterns of the motor coils Mu, Mv, and Mw depending on the states of the phase currents Iu, Iv, and Iw, and each corresponds to a different energized phase. FIG. 4 is a table showing the states of the phase currents Iu, Iv, and Iw in each of the energized phases I to VI.

The phase currents Iu, Iv, and Iw each have six states.
First, there are two states, “source” and “sink”, depending on the direction of the current. The direction of the phase current is determined by which one of the two power transistors directly connected to the corresponding motor coil is turned on.
For example, when the U-phase high-side power transistor 31U is turned on, the U-phase current Iu flows through the U-phase motor coil Mu in the direction of the arrow shown in FIG. The reverse is true when the U-phase low-side power transistor 32U is turned on.
In FIG. 4, the direction when the high-side power transistor is turned on is “source”, and the direction when the low-side power transistor is turned on is “sink”.

Next, there are three states of “increase”, “decrease”, and “fixed” depending on the mode of change in the current level. The phase current level change depends on whether one of the two power transistors directly connected to the corresponding motor coil is kept on during the energized phase or according to either PWM control signals P1 and P2. It depends on whether it is turned on or off.
For example, when the U-phase power transistor 31U or 32U is turned on / off according to the first PWM control signal P1, the U-phase current Iu increases. When the U-phase power transistor 31U or 32U is turned on / off according to the second PWM control signal P2, the U-phase current Iu decreases. When the U-phase power transistor 31U or 32U is maintained in the ON state, the U-phase current Iu is maintained constant.

  The predrive circuit 2 switches the power transistor to be turned on / off every time the energized phase switching signal CP is received. Thereby, the states of the phase currents Iu, Iv, and Iw are switched in the order of I to VI shown in FIG. That is, the phase currents Iu, Iv, and Iw repeat increasing, fixing, and decreasing in the source direction and increasing, fixing, and decreasing in the sink direction, respectively. Between the three phase currents, the phase of the state change is shifted by 120 ° (electrical angle).

FIG. 5 is a waveform diagram of the phase currents Iu, Iv, Iw and BEMFVu, Vv, Vw. FIG. 6 shows an enlarged view of waveforms in energized phases I to IV shown in FIG. A solid line shows a phase current, and a broken line shows BEMF.
The pre-drive circuit 2 turns on / off the high-side power transistor corresponding to the first shaded portion A1 (see FIGS. 5 and 6) according to the first PWM control signal P1. As a result, as indicated by the first hatched portion A1, the phase current increases in a fine step shape in the source direction and is maintained at the highest level (11th level) (see FIG. 6).

  The pre-drive circuit 2 turns on and off the high-side power transistor corresponding to the second shaded portion A2 (see FIGS. 5 and 6) according to the second PWM control signal P2. Thereby, as indicated by the second hatched portion A2, the phase current decreases in a fine step shape in the source direction. When the phase current reaches 0 or when the PWM inhibition signal NPWM rises, the pre-drive circuit 2 maintains the corresponding high-side power transistor in the off state. Thereby, the phase current is interrupted.

  Here, when the energized phase is switched, that is, when a period TP (see FIG. 6) of a constant electrical angle (eg, 37.5 °) has elapsed from the time when the energized phase switching signal CP is received, the PWM inhibition signal NPWM rises and the next energization It remains active until the phase is switched (see below for details). During the period in which the PWM prohibition signal NPWM is maintained active (hereinafter referred to as PWM prohibition period), the predrive circuit 2 maintains the corresponding power transistor in the OFF state regardless of the second PWM control signal P2, and the corresponding phase. Cut off current.

The first shaded portion A1 and the second shaded portion A2 are included in the same energized phase (see FIGS. 5 and 6). For example, in the energized phase I, the U-phase high-side power transistor 31U is turned on / off according to the first PWM control signal P1, the V-phase low-side power transistor 32V is kept on, and the W-phase high-side power transistor 31W is the second PWM. Turns on and off according to control signal P2.
Here, the phase difference between the two set pulse signals SP1 and SP2, that is, the phase difference between the first PWM control signal P1 and the second PWM control signal P2 is maintained substantially equal to 180 °. Therefore, in general, the ON period of the U-phase high-side power transistor 31U in the first hatched portion A1 and the ON period of the W-phase high-side power transistor 31W in the second hatched portion A2 do not overlap.

  In the active period of the first PWM control signal P1, the W-phase current Iw circulates through a free wheel diode connected in parallel to the V-phase motor coil Mv, the V-phase low-side power transistor 32V, and the W-phase low-side power transistor 32W. Therefore, the current I flowing through the current detection resistor R (see FIG. 1) is substantially equal to the U-phase current Iu. That is, the first output signal IC1 (see FIG. 1) of the current comparison unit 13 indicates the difference between the U-phase current Iu and the target value (see FIG. 3) indicated by the increased torque command TQ1. Therefore, the on / off control according to the first PWM control signal P1 increases the U-phase current Iu equal to the target value indicated by the increase torque command TQ1. As a result, the U-phase current Iu increases at a constant rate of change in the source direction (see FIGS. 5 and 6).

  Similarly, during the active period of the second PWM control signal P2, the U-phase current Iu passes through a free wheel diode connected in parallel to the V-phase motor coil Mv, the V-phase low-side power transistor 32V, and the U-phase low-side power transistor 32U. Circulate. Therefore, the current I flowing through the current detection resistor R (see FIG. 1) is substantially equal to the W-phase current Iw. That is, the second output signal IC2 (see FIG. 1) of the current comparison unit 13 indicates the difference between the W-phase current Iw and the target value (see FIG. 3) indicated by the reduced torque command TQ2. Therefore, the on / off control according to the second PWM control signal P2 decreases the W-phase current Iw equally with the target value indicated by the decrease torque command TQ2. As a result, the W-phase current Iw decreases in the source direction at the same rate of change as the rate of change of the U-phase current Iu (see FIGS. 5 and 6).

The V-phase current Iv flows with a constant magnitude in the sink direction (see FIGS. 5 and 6). At that time, the magnitude of the V-phase current Iv is equal to the sum of the magnitudes of the U-phase current Iu and the W-phase current Iw, and particularly equal to the target value indicated by the original torque command TQ (see FIG. 3).
When a certain period of time TP elapses from the start of energized phase I, U-phase current Iu reaches a peak value (target value indicated by original torque command TQ), and W-phase current Iw is interrupted. Thus, commutation from the W phase to the U phase is completed.
The above energization control in energized phase I and the change in energized state due thereto are the same in energized phases III and V.

  The pre-drive circuit 2 turns on and off the low-side power transistor corresponding to the third hatched portion A3 (see FIGS. 5 and 6) according to the first PWM control signal P1. Thereby, as indicated by the third hatched portion A3, the phase current increases in a fine step shape in the sink direction and is maintained at the highest level (11th level) (see FIG. 6).

  The pre-drive circuit 2 turns on and off the low-side power transistor corresponding to the fourth shaded portion A4 (see FIGS. 5 and 6) according to the second PWM control signal P2. As a result, as indicated by the fourth hatched portion A4, the phase current decreases in a fine step shape in the sink direction. When the phase current reaches 0 or when the PWM inhibition signal NPWM rises, the pre-drive circuit 2 maintains the corresponding low-side power transistor in the off state. Thereby, the phase current is interrupted.

The third shaded portion A3 and the fourth shaded portion A4 are included in the same energized phase (see FIGS. 5 and 6). For example, in the energized phase II, the U-phase high-side power transistor 31U is kept on, the V-phase low-side power transistor 32V is turned on / off according to the second PWM control signal P2, and the W-phase low-side power transistor 32W is controlled by the first PWM control. Turns on and off according to signal P1.
Here, the ON period of the V-phase low-side power transistor 32V in the fourth hatched portion A4 and the ON period of the W-phase low-side power transistor 32W in the third hatched portion A3 generally do not overlap.

  In the active period of the first PWM control signal P1, the V-phase current Iv circulates through the free wheel diode connected in parallel to the V-phase high-side power transistor 31V, the U-phase high-side power transistor 31U, and the U-phase motor coil Mu. To do. Therefore, the current I flowing through the current detection resistor R (see FIG. 1) is substantially equal to the W-phase current Iw. That is, the first output signal IC1 (see FIG. 1) of the current comparison unit 13 indicates the difference between the W-phase current Iw and the target value (see FIG. 3) indicated by the increase torque command TQ1. Therefore, the on / off control according to the first PWM control signal P1 increases the W-phase current Iw equal to the target value indicated by the increase torque command TQ1. As a result, the W-phase current Iw increases at a constant rate of change in the sink direction (see FIGS. 5 and 6).

  Similarly, in the active period of the second PWM control signal P2, the W-phase current Iw is connected in parallel to the W-phase high-side power transistor 31W, the freewheel diode, the U-phase high-side power transistor 31U, and the U-phase motor coil. Circulate through Mu. Therefore, the current I flowing through the current detection resistor R (see FIG. 1) is substantially equal to the V-phase current Iv. That is, the second output signal IC2 (see FIG. 1) of the current comparison unit 13 indicates the difference between the V-phase current Iv and the target value (see FIG. 3) indicated by the reduced torque command TQ2. Therefore, the on / off control according to the second PWM control signal P2 reduces the V-phase current Iv to be equal to the target value indicated by the decrease torque command TQ2. As a result, the V-phase current Iv decreases in the sink direction at the same rate of change as the rate of change of the W-phase current Iw (see FIGS. 5 and 6).

The U-phase current Iu flows with a constant magnitude in the source direction (see FIGS. 5 and 6). At that time, the magnitude of the U-phase current Iu is equal to the sum of the magnitudes of the V-phase current Iv and the W-phase current Iw, in particular, equal to the target value indicated by the original torque command TQ (see FIG. 3).
When a certain period TP elapses from the start of energized phase II, V-phase current Iv is cut off and W-phase current Iw reaches a peak value (target value indicated by original torque command TQ). Thus, commutation from the V phase to the W phase is completed.
The above energization control in energization phase II and the change in energization state due thereto are the same in energization phases IV and VI.

  In the above example, using the phase difference between the two set pulse signals SP1 and SP2, the current comparator 13 detects the three phase currents Iu, Iv, and Iw using the common current detection resistor R, respectively. In addition, the current comparison unit 13 may separately detect at least two of the phase currents by using, for example, at least two current detection resistors. At that time, the two set pulse signals SP1 and SP2 may have the same phase, or may be integrated into one set pulse signal.

  Switching the energization of the motor coil changes the magnetic field penetrating the rotor and changes the torque generated by the sensorless motor M. In the sensorless motor driving apparatus according to the first embodiment of the present invention, changes in the phase currents Iu, Iv, and Iw all show a gentle trapezoidal shape as shown in FIG. Accordingly, since the change in the generated torque is gradual, vibration of the sensorless motor M and motor echo noise are sufficiently reduced.

  The BEMF comparison unit 4 (see FIG. 1) includes three comparators 4U, 4V, and 4W. The output signals BCU, BCV, BCW of the comparators 4U, 4V, 4W are the potentials of the three drive terminals U0, V0, W0 of the sensorless motor M and the potential of the midpoint terminal C (hereinafter referred to as the midpoint voltage of the motor coil). Difference).

The BEMF comparison unit 4 may detect the following potential difference separately from the above.
For example, one end of three high resistance resistors is connected at one node to form a Y connection. Further, the other ends of the three resistors are connected to the three drive terminals U0, V0, and W0 of the sensorless motor M, respectively. The BEMF comparison unit 4 may consider the potential of the node between the three resistors as a virtual midpoint voltage and detect a difference from the potential of each of the drive terminals U0, V0, and W0.

The self-excited commutation circuit 5 (see FIG. 1) is based on the three output signals BCU, BCV, BCW of the BEMF comparator 4 and matches the potentials of the drive terminals U0, V0, W0 with the midpoint voltage of the motor coil. Is detected.
The potentials of the drive terminals U0, V0, and W0 are equal to voltages obtained by superimposing BEMFVu, Vv, and Vw on the drive voltage applied from the output circuit 3 (see FIG. 1). On the other hand, when the rotor rotates stably, the BEMFVu, Vv, and Vw are equal to the midpoint voltage of the motor coil, that is, a zero cross is generated in the non-energized motor coil in each energized phase, that is, the motor coil in which the phase current is interrupted. . Accordingly, the zero crosses of BEMF Vu, Vv, and Vw are detected from the coincidence of the respective potentials of the drive terminals U0, V0, and W0 and the midpoint voltage of the motor coil.

The drive terminal for detecting the coincidence of the potential with the midpoint voltage of the motor coil differs for each energized phase. The self-excited commutation circuit 5 determines the mode of the zero cross to be detected next in accordance with the zero cross point information ZCP inputted from the outside.
Here, the difference in the motor coil in which the zero cross occurs (three types of Mu, Mv, and Mw) and the difference in the direction change of the phase current before and after the zero cross (change from “source” to “sink” and vice versa By two types), the zero-cross mode is classified into six types.
In the bottom row of the table shown in FIG. 4, the aspect of the zero cross indicated by the zero cross point information ZCP is shown for each of the energized phases I to VI. Here, the motor coils where the zero cross occurs next are represented by U, V, and W (corresponding to Mu, Mv, and Mw, respectively). Furthermore, the direction change of the phase current before and after the next zero crossing is represented by an upward arrow ↑ and a downward arrow ↓ (corresponding to the change from “source” to “sink” and vice versa, respectively). ).
In FIGS. 5 and 6, BEMFVu, Vv, and Vw induced in the motor coils Mu, Mv, and Mw, respectively, are indicated by broken lines when the rotor rotates stably. Furthermore, the zero-crossing occurrence points (hereinafter referred to as zero-crossing points) due to BEMFVu, Vv, and Vw are symbols (U ↑, U ↓, V ↑, V ↓, W ↑, W ↓) representing the aspect of the zero-crossing shown in FIG. ).

  The self-commutated commutation circuit 5 selects one of the three output signals BCU, BCV, BCW of the BEMF comparison unit 4 based on the aspect of the zero cross determined according to the zero cross point information ZCP, and the rising edge of the selected output signal or One of the falling edges is determined as a detection target. Thus, the zero cross detection accuracy is maintained high.

The self-excited commutation circuit 5 further masks the output signals BCU, BCV, BCW of the BEMF comparison unit 4 according to the two types of mask signals MPWM and DZC. Thereby, the detection accuracy of zero crossing is further improved as follows.
The PWM mask signal MPWM is maintained active by the PWM control circuit 14 for a predetermined time from the rising / falling edge of the first PWM control signal P1. The self-excited commutation circuit 5 keeps the output signals BCU, BCV, BCW of the BEMF comparison unit 4 invalid during the active period. Thereby, the zero cross detection by the self-excited commutation circuit 5 is strong against the ripple accompanying the on / off of the power transistor according to the first PWM control signal P1.

FIG. 6 shows waveforms of the PWM inhibition signal NPWM and the BEMF detection signal DZC. The BEMF detection signal DZC rises when a period TD of a constant electrical angle (for example, 45 °) elapses from the switching point of the energized phase, and remains active until the next switching point of the energized phase (details will be described later). Here, the BEMF detection signal DZC rises later than the PWM inhibition signal NPWM only during a certain electrical angle (for example, 45 ° -37.5 ° = 7.5 °).
During the period in which the BEMF detection signal DZC is maintained active (hereinafter referred to as the BEMF detection period), the self-excited commutation circuit 5 performs zero-cross detection. Since the BEMF detection period is included in the PWM prohibition period, on / off of the power transistor according to the second PWM control signal P2 is prohibited in the BEMF detection period, and the phase current is reliably interrupted by the motor coil that is the detection target of the zero cross. In this way, a constant width (for example, about 15 ° (electrical angle) of 45 ° to about 60 °) of the non-energization period of the motor coil that is the detection target of the zero cross is secured, so that the detection accuracy of the zero cross is maintained high. In particular, since the start of the BEMF detection period is later than the start of the PWM prohibition period, a surge voltage / current such as a flyback voltage accompanying the interruption of the phase current does not prevent the zero cross detection.

The self-excited commutation circuit 5 further generates a self-excited commutation signal SC when the zero cross is detected. However, when no zero cross is detected during the BEMF detection period, the self-excited commutation signal SC is not generated.
7 and 8 show that the first PWM control signal P1, the potential VU0 of the U-phase drive terminal U0, the W-phase drive terminal in the vicinity of the switching point from the first energized phase I to the second energized phase II. 5 is a waveform diagram of a potential VW0 of W0, a midpoint voltage VC of a motor coil, a W-phase output signal BCW of a BEMF comparison unit 4, a PWM mask signal MPWM, a BEMF detection signal DZC, and a self-excited commutation signal SC. FIG. 7 shows a case where a zero cross in the W phase occurs during the BEMF detection period, that is, a case of zero cross edge detection. FIG. 8 shows a case where the zero crossing in the W phase occurs before the BEMF detection period, that is, the case of the zero cross state detection.

  In the PWM inhibition period in the first energized phase I, the on / off control for the U-phase high-side power transistor 31U according to the first PWM control signal P1 is continued, and the U-phase current Iu is maintained. Accordingly, the potential VU0 of the U-phase drive terminal U0 changes to a binary value of the high potential of the power supply terminal 33 (see FIG. 1) and the ground potential in accordance with the first PWM control signal P1. Along with this, the potential VW0 of the W-phase drive terminal W0 and the midpoint voltage VC show similar changes, and particularly include noise at the rise and fall (see the solid and broken lines in FIGS. 7 and 8). Thereby, for a while after the rise / fall of the first PWM control signal P1, the W-phase output signal BCW of the BEMF comparison unit 4 is fine and repeats the rise and fall. These variations generally do not correspond to actual zero crossings.

  The PWM mask signal MPWM is kept active for a certain time TM from the rise / fall of the first PWM control signal P1 (in FIG. 7 and FIG. 8, the low potential state of the PWM mask signal MPWM is set to the active state). ) The self-excited commutation circuit 5 keeps the W-phase output signal BCW of the BEMF comparator 4 invalid during the active period TM. Thereby, the noise accompanying the rise / fall of the first PWM control signal P1 does not prevent the zero cross detection.

  In FIG. 7, during the BEMF detection period, a zero cross occurs in the W phase (see the zero cross point W ↓ shown in FIG. 7). The self-excited commutation circuit 5 accurately detects the zero cross point W ↓, and generates a self-excited commutation signal SC at the time of detection. Here, the self-excited commutation signal SC is a rectangular pulse having a constant short pulse width. The BEMF detection signal DZC falls due to the rising of the self-excited commutation signal SC, and the BEMF detection period ends. On the other hand, the energized phase is switched from I to II (see below for details).

  In FIG. 8, before the BEMF detection period, that is, before the BEMF detection signal DZC rises, a zero cross in the W phase occurs (see the zero cross point ZC shown in FIG. 7). The self-excited commutation circuit 5 detects that the potential VW0 of the W-phase drive terminal W0 is lower than the midpoint voltage VC simultaneously with the rise of the BEMF detection signal DZC (see the point W ↓ shown in FIG. 7). Furthermore, the self-excited commutation signal SC is generated at the time of detection. The BEMF detection signal DZC immediately falls due to the rising of the self-excited commutation signal SC, and ends immediately after the BEMF detection period starts. On the other hand, the energized phase is switched from I to II (see below for details).

The separately excited commutation circuit 6 (see FIG. 1) generates a constant pulse signal, that is, a separately excited commutation signal FC at a predetermined period.
FIG. 9 is a block diagram showing the internal configuration of the separately excited commutation circuit 6. As shown in FIG. The separately excited commutation circuit 6 includes a second oscillation circuit 61 and a separately excited commutation signal generation circuit 62. The second oscillation circuit 61 generates a clock signal CLK having a constant period. The separately excited commutation signal generation circuit 62 generates the separately excited commutation signal FC based on the clock signal CLK and the self-excited commutation signal SC as follows.

FIG. 10 is a timing diagram showing the separately excited commutation signal FC, the count CN1 for the pulse of the clock signal CLK, and the self-excited commutation signal SC. FIG. 11 and FIG. 12 are flowcharts showing the operation of generating the separately excited commutation signal FC by the separately excited commutation signal generating circuit 62.
The separately excited commutation signal generation circuit 62 resets the internal counter (see S1 shown in FIG. 11) at the start of control, and counts the pulses of the clock signal CLK (see S2 shown in FIG. 11).
As shown in FIG. 10, during the period when the self-excited commutation signal SC is not input, the separately-excited commutation signal generation circuit 62 generates the separately-excited commutation signal FC every time the count CN1 reaches the first threshold D. (See S3, S4, and S5 shown in FIG. 11), and resets the count CN1 (see S1 shown in FIG. 11). Here, the separately excited commutation signal FC is a pulse having the same shape as the self-excited commutation signal SC. Further, in a period in which the self-excited commutation signal SC is not input, the period (pulse interval) of the separately-excited commutation signal FC is equal to the time T required to generate the clock signal CLK having the number of pulses equal to the first threshold value D.

When the self-excited commutation signal SC is input, the separately-excited commutation signal generation circuit 62 resets the count CN1 for the pulse of the clock signal CLK even before reaching the first threshold D (shown in FIGS. 10 and 11). S3 and S6 shown in FIG. 12).
Then, the separately excited commutation signal generation circuit 62 replaces the first threshold value D with the second threshold value 6D, and again counts the pulses of the clock signal CLK (see S7 shown in FIG. 12).
Here, the second threshold value 6D is larger than the first threshold value D, and preferably six times the first threshold value D. In addition, it may be 2 times or 20 times, for example.

  When the next self-excited commutation signal SC is input before the count CN1 reaches the second threshold 6D, the other-excited commutation signal generation circuit 62 resets the count CN1, and the pulse of the clock signal CLK starts from 0 again. Count (see S6, S7, S8 shown in FIG. 12). At that time, the threshold value of the count CN1 is maintained at the second threshold value 6D. Thus, when the generation of the self-excited commutation signal SC is repeated and the pulse interval is shorter than the time 6T required for generating the clock signal CLK having the number of pulses equal to the second threshold 6D, the other-excited commutation signal FC is not generated. .

  As shown in FIG. 10, when the count CN1 reaches the second threshold 6D before the next self-excited commutation signal SC is input, the separately-excited commutation signal generation circuit 62 generates the separately-excited commutation signal FC. (See S8 and S9 shown in FIG. 12 and S5 shown in FIG. 11). Further, the loops S1 to S5 shown in FIG. 11 are repeated until the self-excited commutation signal SC is input again. In particular, the pulse interval of the separately excited commutation signal FC is set again to the time T required to generate the clock signal CLK having the number of pulses equal to the first threshold value D (see FIG. 10).

The count unit 7 (see FIG. 1) includes a selection circuit 71 and a count circuit 72.
The selection circuit 71 selects the self-commutated commutation signal SC or the separately-excited commutation signal FC, whichever is input earlier, as the commutation signal CS, and sends it to the count circuit 72 and the energized phase switching circuit 8.
The energized phase switching circuit 8 (see FIG. 1) sends an energized phase switching signal CP to the pre-drive circuit 2 in synchronization with the commutation signal CS, and further sends zero-cross point information ZCP to the self-excited commutation circuit 5. Thereby, the energized phase is switched in synchronization with the commutation signal CS.

The count circuit 72 measures the interval of the commutation signal CS and generates the position signal PS based on the interval. Further, based on the position signal PS, a PWM inhibition signal NPWM and a BEMF detection signal DZC are generated.
FIG. 13 is a block diagram showing the internal configuration of the count circuit 72. As shown in FIG. FIG. 14 is a timing chart of the commutation signal CS, the internal signals CN2, RTC, DV of the count circuit 72, and the three signals PS, NPWM, DZC sent from the count circuit 72.

The count circuit 72 includes a 60 ° section count circuit 721, a data holding circuit 722, a division circuit 723, and a position signal generation circuit 724.
The 60 ° section count circuit 721 resets the count CN2 of the internal counter when the commutation signal CS is input, and counts the pulses of the internal clock until the next commutation signal CS is input.

The data holding circuit 722 reads the count CN2 of the internal counter from the 60 ° interval count circuit 721 in synchronization with the commutation signal CS. Further, the read count CN2 is compared with the number of pulses corresponding to the pulse interval T of the separately excited commutation signal FC. Thereby, the smaller one of the former and the latter is held as the holding data RTC until the next commutation signal CS is input.
For example, as shown in FIG. 14, when the pulse interval of the commutation signal CS changes as T1, T2, T3, T4, T5, T6, and they are all shorter than the pulse interval T of the separately excited commutation signal FC. Suppose. The retained data RTC of the data retaining circuit 722 is then equal to the count D1 corresponding to the pulse interval T1 during the pulse interval T2. Similarly, the retained data RTC is equal to the counts D2, D3, D4, and D5 corresponding to the pulse intervals T2, T3, T4, and T5, respectively, in the pulse intervals T3, T4, T5, and T6.
For example, when the pulse interval T1 of the commutation signal CS is longer than the pulse interval T of the separately excited commutation signal FC, the retained data RTC of the data holding circuit 722 is the pulse interval T of the separately excited commutation signal FC during the period of the pulse interval T2. Is equal to the number of pulses corresponding to.

The division circuit 723 divides the held data RTC of the data holding circuit 722 by 16, for example, when the commutation signal CS is input. Here, the division number may be another number, for example, 8 or 32.
The dividing circuit 723 includes, for example, a 4-bit counter inside. At that time, the dividing circuit 723 increments the count of the 4-bit counter by one every time the internal clock pulse having the same period as the internal clock of the 60 ° section counting circuit 721 is counted by 1/16 of the holding data RTC. The count is generated as a 4-bit parallel signal DV (see FIG. 14).
According to the 4-bit state of the parallel signal DV, the pulse interval of the commutation signal CS is substantially divided into 16 by 1/16 of the pulse interval immediately before it. For example, the period of the pulse interval T2 of the commutation signal CS is divided by 1/16 of the pulse interval T1. Similarly, the periods of the pulse intervals T3, T4, T5, and T6 are each divided into 1/16 of the pulse intervals T2, T3, T4, and T5.
However, when the sensorless motor M is started, the dividing circuit 723 replaces the internal clock pulse with 1/16 of the holding data RTC, and 1/16 of the number of pulses corresponding to the pulse interval T of the separately excited commutation signal FC. Each time it counts, the 4-bit counter is incremented by one. Thus, the period until the next commutation signal CS is input is divided into 1/16 of the pulse interval T of the separately excited commutation signal FC.

  The position signal generation circuit 724 reads the 4-bit parallel signal DV from the division circuit 723, and among the 16 pulse waves 0 to 15 of the position signal PS, a pulse having a number equal to the numerical value represented by the 4 bits of the parallel signal DV Raise the wave (see Figure 14). In particular, since the leading pulse wave (number 0) is synchronized with the commutation signal CS, a series of pulses of numbers 0 to 15 divide the pulse interval of the commutation signal CS into 16 sections. Since the pulse interval of the commutation signal CS determines one energized phase period (60 ° (electrical angle) period), the interval per pulse of the position signal PS corresponds to 3.75 ° (electrical angle). Thus, each of the pulses numbered 0 to 15 indicates the estimated position of the rotor of the sensorless motor M at intervals of 3.75 ° (electrical angle) (see FIGS. 3 and 14).

Further, the position signal generation circuit 724 raises the PWM inhibition signal NPWM in synchronization with the rising of the pulse of the fixed number of the position signal PS, for example, the pulse of the number 10 (see FIG. 14). As a result, the time TP from the time when the commutation signal CS is generated, that is, the time when the energized phase is switched to the time when the PWM inhibition period starts is set to a constant electrical angle (for example, 37.5 °).
Here, the period TP may be a value different from 37.5 °, for example, 30 °. Furthermore, the period TP may be set longer than the period of change in level of the decrease torque command TQ2 (see FIG. 3).

After the rise of the PWM inhibition signal NPWM, the position signal generation circuit 724 raises the BEMF detection signal DZC in synchronization with the rising of the pulse of the fixed number of the position signal PS, for example, the pulse of the number 12 (see FIG. 14). Thereby, the period TD from the generation time of the commutation signal CS, that is, the switching time of the energized phase to the start time of the BEMF detection period, is set to a constant electrical angle (for example, 45 °) period. In particular, the start time of the BEMF detection period is delayed by a certain electrical angle period (for example, 7.5 °) from the start time of the PWM inhibition period.
Here, the period TD may be a value different from 45 °, for example, 40 °. That is, the period from the start time of the PWM prohibition period to the start time of the BEMF detection period may be set to a value (for example, 10 °) different from 7.5 °.

  When the rotor of the sensorless motor M rotates stably, the self-excited commutation circuit 5 can stably repeat the accurate detection of zero crossing as described above. Here, the extended pulse interval 6T of the separately excited commutation signal FC is set sufficiently longer than the pulse interval of the self-excited commutation signal SC. Accordingly, the count unit 7 continues to send the self-excited commutation signal SC as the commutation signal CS. Therefore, the switching of the energized phase is accurately synchronized with the rotation of the rotor. In this way, the rotational stability of the rotor is kept high.

In the sensorless motor driving apparatus according to the first embodiment of the present invention, as described above, the PWM inhibition period is set in a certain period immediately before the zero cross point. Thereby, in each motor coil, the phase current reaches zero earlier than BEMF.
Further, the energized phase is switched in synchronization with the commutation signal, and in particular, the PWM inhibition period and the BEMF detection period end simultaneously. Thereby, when the zero cross is detected, the non-energized state of the motor coil is quickly released, and the phase current increases rapidly.
FIG. 15 is a waveform diagram of the phase current and BEMF for the same motor coil when the rotor of the sensorless motor M rotates stably. In FIG. 15, the solid line indicates the phase current, and the broken line indicates BEMF. As shown in FIG. 15, the phase current reaches zero prior to BEMF by setting the PWM inhibition period TP. On the other hand, at the zero cross point ZC of BEMF, the source current or the sink current immediately increases immediately after the detection. As a result, the phase current waveform is pseudo-advanced from the BEMF waveform (see the arrow shown in FIG. 15). A rapid rise in phase current represents a rapid increase in generated torque.

  Thus, the sensorless motor driving apparatus according to Embodiment 1 of the present invention secures a PWM prohibition period, that is, a non-energization period of the motor coil, when the motor echo noise is suppressed by a gradual change of the phase current. Detection is possible. Furthermore, since the phase current quickly rises simultaneously with the detection of zero crossing, that is, switching of the energized phase, the generated torque can be sufficiently increased. As a result, the drive control of the sensorless motor is particularly strong against load fluctuations.

The sensorless motor driving apparatus according to the first embodiment of the present invention further starts the sensorless motor M by the following separately excited commutation control. In particular, by using the above-described configuration, switching from the separately excited commutation control to the self-excited commutation control is realized quickly and reliably as follows.
16, 17, and 18 are timing charts showing the separately excited commutation signal FC, the self-excited commutation signal SC, the commutation signal CS, and the phase currents Iu, Iv, and Iw when the sensorless motor M is started. It is. 16 to 18, the BEMF detection period is indicated by a hatched portion.

From the start time of the sensorless motor M to the first detection time of zero crossing, that is, from the generation time of the first self-excited commutation signal SC (see time TC shown in FIGS. Select FC as the commutation signal CS. Thereby, the switching period of the energized phase is equal to the period T of the separately excited commutation signal FC.
Furthermore, the pulse width of the position signal PS is equal to 1/16 of the period T of the separately excited commutation signal FC (see FIG. 14). Accordingly, the waveforms of the phase currents Iu, Iv, and Iw are set based on the period T of the separately excited commutation signal FC. For example, the period of the phase current is equal to 6 times the period T of the separately excited commutation signal FC, and the increase / decrease period of the phase current is equal to 37.5 ° / 60 ° × T.
In addition, the period TD from the generation time of the separately excited commutation signal FC to the start time of the BEMF detection period (see the hatched portion shown in FIGS. 16 to 18) is constant based on the period T of the separately excited commutation signal FC. Determined (for example, TD = 45 ° / 60 ° × T). The same applies to the PWM inhibition period.

  16-18, the first self-excited commutation signal SC is generated at time TC. Since the self-excited commutation signal SC is always generated during the BEMF detection period, the self-excited commutation signal SC and the immediately preceding other-excited commutation signal FC (the other-excited commutation signal at time T0 shown in FIGS. 16 to 18). FC) is always shorter than the pulse interval T of the separately excited commutation signal FC. Therefore, whenever the self-excited commutation signal SC is generated, the self-excited commutation signal SC is first input to the selection circuit 71 and selected as the commutation signal CS. Thereby, the energized phase is switched according to the self-excited commutation signal SC instead of the separately-excited commutation signal FC. Thus, the separately excited commutation control can be quickly and reliably switched to the self-excited commutation control.

At the generation time TC of the self-excited commutation signal SC, the separately-excited commutation circuit 6 extends the period (pulse interval) of the separately-excited commutation signal FC by 6 times (see FIGS. 10 and 12).
Further, in the energized phase immediately after the time TC, the pulse width of the position signal PS is equal to 1/16 of the time T1 of the immediately preceding energized phase (see FIG. 14). Therefore, the waveforms of the phase currents Iu, Iv, and Iw are set based on the immediately previous energized phase time T1. For example, the phase current increase / decrease period is equal to 37.5 ° / 60 ° × T1.
In addition, a period TD1 from time TC to the start point of the BEMF detection period is set based on the time T1 of the immediately previous energized phase (for example, TD1 = 45 ° / 60 ° × T1). The same applies to the PWM inhibition period.

After time TC, when the generation of the self-excited commutation signal SC is repeated and their pulse interval is shorter than the extended pulse interval 6T of the separately excited commutation signal FC, the commutation signal CS is self-excited commutation signal SC. (See FIG. 16 and FIG. 17).
When the pulse interval T2 of the self-excited commutation signal SC is shorter than the original pulse interval T of the separately excited commutation signal FC (T2 <T) as shown in FIG. 16, for example, the pulse of the position signal PS in the next energized phase The width is equal to 1/16 of the pulse interval T2 of the self-excited commutation signal SC (see FIG. 14). Accordingly, the waveforms of the phase currents Iu, Iv, and Iw are set based on the pulse interval T2 of the self-excited commutation signal SC. For example, the phase current increase / decrease period is equal to 37.5 ° / 60 ° × T2.
Further, a period TD2 from the switching time of the energized phase to the start time of the BEMF detection period is set based on the pulse interval T2 of the self-excited commutation signal SC (for example, TD2 = 45 ° / 60 ° × T2). The same applies to the PWM inhibition period.

For example, as shown in FIG. 17, the pulse interval T2 of the self-excited commutation signal SC is longer than the original pulse interval T of the separately-excited commutation signal FC and is longer than the extended pulse interval 6T of the separately-excited commutation signal FC. When it is short (T <T2 <6T), in the next energized phase, the pulse width of the position signal PS is set equal to 1/16 of the original pulse interval T of the separately excited commutation signal FC (see FIG. 14). Therefore, the waveforms of the phase currents Iu, Iv, and Iw are set based on the original pulse interval T of the separately excited commutation signal FC. For example, the phase current increase / decrease period is equal to 37.5 ° / 60 ° × T.
Further, a period TD from the switching time of the energized phase to the start time of the BEMF detection period is set based on the pulse interval T of the separately excited commutation signal FC (for example, TD = 45 ° / 60 ° × T). The same applies to the PWM inhibition period.

For example, as shown in FIG. 18, when a new self-excited commutation signal SC is not generated from time TC until the elapse of the extended period 6T of the other-excited commutation signal FC, the separately-excited commutation signal FC is Selected as the next commutation signal CS. That is, the switching of the next energized phase is executed by generating the separately excited commutation signal FC (see time TC2 shown in FIG. 18).
In the energized phase, the pulse width of the position signal PS is set equal to 1/16 of the original pulse interval T of the separately excited commutation signal FC (see FIG. 14). Therefore, the waveforms of the phase currents Iu, Iv, and Iw are set based on the original pulse interval T of the separately excited commutation signal FC. For example, the phase current increase / decrease period is equal to 37.5 ° / 60 ° × T.
Further, a period TD from the switching time of the energized phase to the start time of the BEMF detection period is set based on the pulse interval T of the separately excited commutation signal FC (for example, TD = 45 ° / 60 ° × T). The same applies to the PWM inhibition period.
In addition, the separately excited commutation circuit 6 returns the period of the separately excited commutation signal FC to the original period T at time TC2. Thereafter, the separately excited commutation control is continued until the self-excited commutation signal SC is generated again.

The situation as shown in FIGS. 17 and 18 is likely to occur during a period when the rotational speed of the rotor is small, for example, when the sensorless motor M is started. In such a situation, since the BEMF level is generally low, the zero-cross detection accuracy decreases. In order to maintain high zero-cross detection accuracy, it is preferable to extend the BEMF detection period.
In the sensorless motor driving apparatus according to Embodiment 1 of the present invention, as shown in FIGS. 17 and 18, a long BEMF detection period is ensured. Accordingly, the detection accuracy of the zero cross is maintained high, so that the sensorless motor can be started particularly smoothly and accurately.
Furthermore, the pulse width of the position signal PS does not exceed 1/16 of the original pulse interval T of the separately excited commutation signal FC, regardless of the time of the immediately preceding energized phase. Thereby, the rising of the phase current is maintained at a certain speed or higher when the energized phase is switched. Therefore, the generated torque is sufficiently large. Particularly when the sensorless motor M is started, the starting torque is large, so that the starting time can be shortened and the starting control is strong against load fluctuations.

  Furthermore, in the sensorless motor driving apparatus according to the first embodiment of the present invention, when the detection of zero crossing is not repeated, thereby failing in switching from the separately excited commutation control to the self-excited commutation control, as shown in FIG. The separately excited commutation control similar to that immediately after the start of the sensorless motor M is resumed promptly. That is, unlike the retry in the conventional sensorless motor driving device, for example, it is not necessary to determine the rotational speed of the rotor or to wait for a retry command from an external microprocessor. In this way, the sensorless motor can be started more smoothly.

When the sensorless motor M is used as a spindle motor of a CD / DVD drive or HDD of a portable information device, for example, the sensorless motor M being driven is likely to receive excessive vibration / impact from the outside. Therefore, it is assumed that the rotor receives sudden torque due to excessive vibration / impact, whereby the zero cross point of BEMF suddenly shifts greatly and the generation of the self-excited commutation signal SC suddenly stops.
In the sensorless motor driving apparatus according to the first embodiment of the present invention, preferably, the separately excited commutation circuit 6 constantly generates the separately excited commutation signal FC throughout the driving period of the sensorless motor M. Thereby, even when the generation of the self-excited commutation signal SC suddenly stops as described above, the selection circuit 71 can immediately select the separately-excited commutation signal FC as the commutation signal CS. That is, the self-excited commutation control is promptly switched to the separately excited commutation control. Further, the separately excited commutation control can be quickly and reliably switched to the self-excited commutation control as described above.
Thus, the control by the sensorless motor driving apparatus according to the first embodiment of the present invention is strong against external vibration / impact.

  In the sensorless motor driving apparatus according to the first embodiment of the present invention, the PWM control unit 1 directly adjusts the phase currents Iu, Iv, and Iw by current drive control as described above. In addition, the PWM control unit 1 controls the phase currents Iu, Iv, and Iw by voltage drive control, that is, by controlling the drive voltage applied from the output circuit 3 to each of the three drive terminals U0, V0, and W0 of the sensorless motor M. It may be adjusted indirectly.

  In the sensorless motor driving apparatus according to the first embodiment of the present invention, the PWM control unit 1 maintains the phase currents Iu, Iv, and Iw in gentle waveforms as described above. In addition, the PWM controller 1 may maintain the phase currents Iu, Iv, and Iw in a rectangular wave as in the conventional sensorless motor driving device (see, for example, FIG. 28). Even in such a case, the quickness and certainty of switching from the separately excited commutation control to the self-excited commutation control are not impaired, so that the sensorless motor can be quickly and surely started.

<< Embodiment 2 >>
The sensorless motor driving apparatus according to the second embodiment of the present invention is driven by, for example, a three-phase (U-phase, V-phase, W-phase) sensorless motor, similar to the driving apparatus according to the first embodiment. The components of the sensorless motor driving device are the same as those of the driving device according to the first embodiment except for the count circuit 72 (see FIG. 1). For the similar components, the description in the first embodiment and the cited drawings are incorporated.

The sensorless motor driving device according to the second embodiment of the present invention differs from the driving device according to the first embodiment in the internal configuration of the count circuit 72. FIG. 19 is a block diagram showing an internal configuration of the count circuit 72 in the sensorless motor driving apparatus according to the second embodiment of the present invention. FIG. 20 is a timing chart of the commutation signal CS, the 4-bit parallel signal DV, the position signal PS, the PWM inhibition signal NPWM, the trigger signal TR, and the BEMF detection signal DZC.
In FIG. 19 and FIG. 20, the same reference numerals as those shown in FIG. 13 and FIG. 14 are given to the same constituent elements as those shown in FIG. 13 and FIG. The explanation in is used.

The count circuit 72 includes a fixed time count circuit 725 in addition to the 60 ° interval count circuit 721, the data holding circuit 722, the division circuit 723, and the position signal generation circuit 724A.
The fixed time count circuit 725 reads the 4-bit parallel signal DV from the dividing circuit 723 and measures the fixed time ΔT from the rising edge of the pulse of the fixed number of the position signal PS, for example, the pulse of the number 10. The fixed time count circuit 725 further sends a trigger signal TR to the position signal generation circuit 724A when the fixed time ΔT has elapsed.
Here, the measurement start time point by the constant time counting circuit 725 may be the rising time point of a pulse having a number different from the number 10 of the position signal PS. However, the measurement start time is set to the rise time of the PWM inhibition signal NPWM.

  The position signal generation circuit 724A reads the 4-bit parallel signal DV from the division circuit 723, and among the 16 pulse waves 0 to 15 of the position signal PS, a pulse having a number equal to the numerical value represented by the 4 bits of the parallel signal DV Raise the wave (see Figure 20). In particular, since the leading pulse wave (number 0) is synchronized with the commutation signal CS, a series of pulses of numbers 0 to 15 divide the pulse interval of the commutation signal CS into 16 sections. Since the pulse interval of the commutation signal CS determines one energized phase period (60 ° (electrical angle) period), the interval per pulse of the position signal PS corresponds to 3.75 ° (electrical angle). Thus, each of the pulses of numbers 0 to 15 indicate the estimated position of the rotor of the sensorless motor M at intervals of 3.75 ° (electrical angle) (see FIGS. 3 and 19).

Further, the position signal generation circuit 724A raises the PWM inhibition signal NPWM in synchronization with the rising of a pulse of a fixed number of the position signal PS, for example, the pulse of number 10 (see FIG. 20). As a result, the time TP from the time when the commutation signal CS is generated, that is, the time when the energized phase is switched to the time when the PWM inhibition period starts is set to a constant electrical angle (for example, 37.5 °).
Here, the period TP may be a value different from 37.5 °, for example, 30 °. Furthermore, the period TP may be set longer than the period of change in level of the decrease torque command TQ2 (see FIG. 3).

The position signal generation circuit 724A raises the BEMF detection signal DZC in synchronization with the rising edge of the position signal PS having a certain number, for example, the number 12 pulse or the trigger signal TR which is input earlier (FIG. 20). reference).
For example, as shown in the left half of FIG. 20, when the rotational speed of the rotor is small, the pulse interval T1 of the commutation signal CS is long, so that the trigger signal TR is generated prior to the number 12 pulse of the position signal PS. At that time, the BEMF detection signal DZC rises in synchronization with the rise of the trigger signal TR. Therefore, the period from the switching point of the energized phase to the start point of the BEMF detection period is shorter than the period TD of 37.5 ° (electrical angle).

On the other hand, for example, as shown in the right half of FIG. 20, when the number of rotations of the rotor increases sufficiently, the pulse interval T2 of the commutation signal CS is sufficiently short, so that the number 12 pulse of the position signal PS is the trigger signal. Generated before TR. At that time, the BEMF detection signal DZC rises in synchronism with the rise of the number 12 pulse of the position signal PS. Therefore, the period from the switching point of the energized phase to the start point of the BEMF detection period coincides with the period TD of 37.5 ° (electrical angle).
Thus, regardless of the number of rotations of the rotor, the time from the start of the PWM inhibition period to the start of the BEMF detection period is kept sufficiently small to avoid erroneous detection of zero cross due to, for example, a flyback voltage. That is, the BEMF detection period starts promptly. As a result, the detection accuracy of the zero cross is kept high regardless of the rotational speed of the rotor.

<< Embodiment 3 >>
The sensorless motor driving apparatus according to the third embodiment of the present invention is driven by, for example, a three-phase (U-phase, V-phase, W-phase) sensorless motor, similar to the driving apparatus according to the first embodiment. The components of the sensorless motor driving device are the same as those of the driving device according to the first embodiment except for the separately excited commutation circuit 6 (see FIG. 1). For the similar components, the description in the first embodiment and the cited drawings are incorporated.

  The sensorless motor driving device according to the third embodiment of the present invention differs from the driving device according to the first embodiment in the internal configuration of the separately excited commutation circuit 6. FIG. 21 is a block diagram showing an internal configuration of the separately excited commutation circuit 6 in the sensorless motor driving apparatus according to Embodiment 3 of the present invention. In FIG. 21, the same components as those shown in FIG. 9 are denoted by the same reference numerals as those shown in FIG. 9, and the description of the first embodiment is used for the description thereof.

The separately excited commutation circuit 6 includes a second oscillation circuit 61, a separately excited commutation signal generation circuit 62A, and a separately excited commutation signal control circuit 63.
The separately excited commutation signal generation circuit 62A and the separately excited commutation signal control circuit 63 generate the separately excited commutation signal FC based on the clock signal CLK and the self-excited commutation signal SC as follows.
FIG. 22 shows two signals PA and PB generated by the separately excited commutation signal generation circuit 62A, the first count CN1A and the second count CN1B for the pulse of the clock signal CLK, the separately excited commutation signal FC, and It is a timing diagram which shows the self-excited commutation signal SC. 23, 24, and 25 are flowcharts showing the operation of generating the separately excited commutation signal FC by the separately excited commutation signal generating circuit 62A and the separately excited commutation signal control circuit 63. FIG.

The separately excited commutation signal generation circuit 62A has two counters therein. At the start of control (time TS shown in FIG. 22), the separately excited commutation signal generation circuit 62A first resets both counters (see S11 shown in FIG. 23), and the first counter generates a pulse of the clock signal CLK. Count (see S12 shown in FIG. 23).
As shown in FIG. 22, when the count CN1A of the first counter reaches the first threshold D before the self-excited commutation signal SC is input, the separately excited commutation signal generation circuit 62A generates the first signal PA. (See S13 and S14 shown in FIG. 23). Further, the separately excited commutation signal control circuit 63 sends the first signal PA as the separately excited commutation signal FC (see S15 shown in FIG. 23).
Here, the first signal PA is a pulse having the same shape as the self-excited commutation signal SC.

In the period when the self-excited commutation signal SC is not input, the separately-excited commutation signal generation circuit 62A and the separately-excited commutation signal control circuit 63 count the pulses of the first signal PA (see S16 shown in FIG. 24). As shown in FIG. 22, in a period when the self-excited commutation signal SC is not input and the number of pulses of the first signal PA is a fixed number, for example, less than 6, the count CN1A of the first counter is the first Each time the threshold value D is reached, the reset of the count CN1A of the first counter, the generation of the first signal PA, and the transmission of the first signal PA as the other-excited commutation signal FC are repeated (FIG. 23 and FIG. 23). (See S11 to S16 shown in FIG. 24).
Here, in a period in which the self-excited commutation signal SC is not input, the period (pulse interval) of the first signal PA is equal to the time TA required to generate the clock signal CLK having the number of pulses equal to the first threshold value D.

As shown in FIG. 22, when the number of pulses of the first signal PA reaches a certain number, for example, 6 before the self-excited commutation signal SC is input, the separately-excited commutation signal generation circuit 62A sets both counters. Reset and count the pulses of the clock signal CLK in the second counter together with the first counter (see S17 and S18 shown in FIG. 24).
As shown in FIG. 22, in the period when the self-excited commutation signal SC is not input, the separately excited commutation signal generation circuit 62A generates the second signal every time the count CN1B of the second counter reaches the second threshold 2D. PB is generated, and the count CN1B of the second counter is reset (see S17, S18, S19, and S20 shown in FIG. 24). Further, during the period from the start of counting by the second counter to the input time of the self-excited commutation signal SC, the separately-excited commutation signal control circuit 63 replaces the first signal PA with the second signal PB. It is sent as an excitation commutation signal FC (see S21 shown in FIG. 24).

  Here, like the first signal PA, the second signal PB is a pulse having the same shape as the self-excited commutation signal SC. Furthermore, the cycle (pulse interval) of the second signal PB is equal to the time TB required to generate the clock signal CLK having the number of pulses equal to the second threshold 2D. The second threshold value 2D is larger than the first threshold value D, and preferably doubled. At that time, the period TB of the second signal PB is longer than the period TA of the first signal PA (TB> TA), and is preferably twice the period TA of the first signal PA.

As shown in FIG. 22, when the self-excited commutation signal SC is input, the separately-excited commutation signal generation circuit 62A sets two counts CN1A and CN1B for the pulse of the clock signal CLK to the respective threshold D and It is reset even before reaching 2D (see S13 shown in FIG. 23, S19 shown in FIG. 24, and S22 shown in FIG. 25).
The separately excited commutation signal generation circuit 62A then replaces the first threshold D with the third threshold for the first counter. Here, the third threshold value is larger than both the first threshold value D and the second threshold value 2D, and is preferably six times the first threshold value D. In addition, it may be 20 times, for example. The third threshold is more preferably an integer multiple of the second threshold 2D.
The separately excited commutation signal generation circuit 62A then counts the pulses of the clock signal CLK again by each of the two counters (see S23 shown in FIGS. 22 and 25).
On the other hand, during the period when the third threshold 6D is set in the first counter, the separately excited commutation signal control circuit 63 selects the first signal PA as the separately excited commutation signal FC instead of the second signal PB. To do. That is, as shown in FIG. 22, the second signal PB is not transmitted as the separately excited commutation signal FC.

When the next self-excited commutation signal SC is input before the count CN1A of the first counter reaches the third threshold 6D, the separately-excited commutation signal generation circuit 62A calculates the counts CN1A and CN1B of the two counters. Reset is performed, and the pulses of the clock signal CLK are counted again from 0 (see S22, S23, and S24 shown in FIG. 25). At that time, the threshold value of the count CN1A of the first counter is maintained at the third threshold value 6D. Further, the separately excited commutation signal control circuit 63 maintains the selection target as the separately excited commutation signal FC at the first signal PA.
Thus, when the generation of the self-excited commutation signal SC is repeated and the pulse interval is shorter than the time 6TA required to generate the clock signal CLK having the number of pulses equal to the third threshold 6D, the other-excited commutation signal FC is generated. Not.

As shown in FIG. 22, when the count CN1A of the first counter reaches the third threshold 6D before the next self-excited commutation signal SC is input, the separately excited commutation signal generation circuit 62A The signal PA is generated. Further, the separately excited commutation signal control circuit 63 transmits the first signal PA as the separately excited commutation signal FC (see S25 and S26 shown in FIG. 25).
The separately excited commutation signal generation circuit 62A then replaces the third threshold 6D with the original first threshold D for the first counter. Further, the separately excited commutation signal control circuit 63 selects the second signal PB as the separately excited commutation signal FC instead of the first signal PA.
Then, the loops S17 to S21 shown in FIG. 24 are repeated until the self-excited commutation signal SC is input again. In particular, the cycle of the separately excited commutation signal FC is set again to the cycle TB of the second signal PB (see FIG. 22).

The sensorless motor driving apparatus according to Embodiment 3 of the present invention starts the sensorless motor M by the following separately excited commutation control. In particular, by using the above-described other-excited commutation circuit 6, switching from the other-excited commutation control to the self-excited commutation control is realized quickly and reliably regardless of the load weight as follows.
FIG. 26 is a timing chart showing the separately excited commutation signal FC, the self-excited commutation signal SC, the commutation signal CS, and the phase currents Iu, Iv, and Iw when the sensorless motor M is started. In FIG. 26, the BEMF detection period is indicated by hatching.

From the start time TS of the sensorless motor M to the first detection time of zero crossing, that is, from the generation time of the first self-excited commutation signal SC (see time TC shown in FIG. 26), the selection circuit 71 performs the separately excited commutation signal FC. Is selected as the commutation signal CS. In that period, first, the first signal PA is sent as the separately excited commutation signal FC, so that the switching phase of the energized phase is the period TA of the first signal PA (hereinafter, the short period TA of the separately excited commutation signal FC). It is equal to).
Furthermore, the pulse width of the position signal PS is equal to 1/16 of the short period TA of the separately excited commutation signal FC (see FIG. 14). Therefore, the waveforms of the phase currents Iu, Iv, and Iw are set based on the short cycle TA of the separately excited commutation signal FC. For example, the period of the phase current is equal to 6 times the short period TA of the separately excited commutation signal FC, and the increase / decrease period of the phase current is equal to 37.5 ° / 60 ° × TA.
In addition, the period TD from the generation point of the separately excited commutation signal FC to the start point of the BEMF detection period (see the shaded area shown in FIG. 26) is determined to be constant based on the short period TA of the separately excited commutation signal FC. (For example, TD = 45 ° / 60 ° × TA). The same applies to the PWM inhibition period.

  When the self-excited commutation signal SC is generated before the self-excited commutation signal FC is continuously generated six times from the start point TS of the sensorless motor M, the energized phase is separately excited as in the first embodiment. Instead of the commutation signal FC, switching is performed according to the self-excited commutation signal SC. In particular, once the self-excited commutation signal SC is generated, the period TA of the first signal PA is extended six times. On the other hand, the second signal PB is not yet generated. Therefore, as in the first embodiment, a long BEMF detection period is ensured (see FIG. 17). As a result, the zero-cross detection accuracy is maintained high, so that the separately excited commutation control can be quickly and reliably switched to the self-excited commutation control. Thus, the start of the sensorless motor can be executed smoothly and accurately.

As shown in FIG. 26, when the separately excited commutation signal FC is continuously generated six times before the self-excited commutation signal SC is generated from the start time TS of the sensorless motor M, the separately excited commutation circuit is generated. 6 sends the second signal PB as a separately excited commutation signal FC. Therefore, the period of the separately excited commutation signal FC is extended to the period TB of the second signal PB (hereinafter referred to as the long period TB of the separately excited commutation signal FC).
On the other hand, the pulse width of the position signal PS is maintained equal to 1/16 of the short period TA of the separately excited commutation signal FC, as in the previous value. Therefore, the waveforms of the phase currents Iu, Iv, and Iw, and the period TD from the switching point of the energized phase to the start point of the BEMF detection period are maintained equal to those up to that time. The same applies to the PWM inhibition period.

When the load of the sensorless motor M is relatively light, the separately-excited commutation control by the separately-excited commutation signal FC of the short cycle TA quickly increases the rotational speed of the rotor. However, when the load of the sensorless motor M is relatively heavy, the separately-excited commutation control by the separately-excited commutation signal FC of the short cycle TA is difficult to increase the rotational speed of the rotor. Accordingly, it is difficult to detect the zero cross with high accuracy.
In the sensorless motor driving apparatus according to the third embodiment of the present invention, as described above, when the zero crossing is not detected for a certain period in the separately excited commutation control by the separately excited commutation signal FC of the short cycle TA, the short circuit of the separately excited commutation signal FC is short. Period TA is extended to long period TB. Thereby, as shown in FIG. 26, the BEMF detection period is extended. Accordingly, the zero cross detection accuracy is improved.
Thus, the sensorless motor driving apparatus according to Embodiment 3 of the present invention can realize a quick and reliable start-up of the sensorless motor regardless of the load weight.
At that time, the pulse width of the position signal PS does not exceed 1/16 of the short period TA of the separately excited commutation signal FC. As a result, when the energized phase is switched, the rising of the phase current is maintained at a certain speed or higher. Therefore, since the starting torque is large, the starting time can be shortened and the starting control is strong against load fluctuation.

  In FIG. 26, the first self-excited commutation signal SC is generated at time TC. As in the first embodiment, whenever the self-excited commutation signal SC is generated, the self-excited commutation signal SC is first input to the selection circuit 71 and selected as the commutation signal CS. Thereby, the energized phase is switched according to the self-excited commutation signal SC instead of the separately-excited commutation signal FC. Thus, the separately excited commutation control can be quickly and reliably switched to the self-excited commutation control.

At the generation time TC of the self-excited commutation signal SC, the separately-excited commutation circuit 6 extends the period (pulse interval) of the separately-excited commutation signal FC to 6 times the short period TA.
Further, in the energized phase immediately after the time TC, the pulse width of the position signal PS, the waveforms of the phase currents Iu, Iv, and Iw, and the period TD1 from the time TC to the start of the BEMF detection period are the time T1 of the immediately previous energized phase. Or the other excitation commutation signal FC is set based on the shorter one of the short periods TA. The same applies to the PWM inhibition period.
Thereby, as shown in FIG. 26, the BEMF detection period is extended. Accordingly, the zero cross detection accuracy is improved.
At that time, the pulse width of the position signal PS does not exceed 1/16 of the short period TA of the separately excited commutation signal FC. As a result, when the energized phase is switched, the rising of the phase current is maintained at a certain speed or higher. Therefore, since the starting torque is large, the starting time can be shortened and the starting control is strong against load fluctuation.

When the generation of the self-excited commutation signal SC is repeated after the time TC and the pulse interval is shorter than 6 times the short period TA of the other-excited commutation signal FC, the commutation signal is the same as in the first embodiment. CS is synchronized with the self-excited commutation signal SC (see FIGS. 16 and 17).
On the other hand, as shown in FIG. 26, when a new self-excited commutation signal SC is not generated from the time TC until the time when six times the short cycle TA of the other-excited commutation signal FC has elapsed, FC is selected as the next commutation signal CS. In other words, the next energized phase is switched by generating the separately excited commutation signal FC (see time TC2 shown in FIG. 26).
In the energized phase, the pulse width of the position signal PS is set equal to 1/16 of the short period TA of the separately excited commutation signal FC. Accordingly, the waveforms of the phase currents Iu, Iv, Iw, and the period TD from the switching time TC2 of the energized phase to the start time of the BEMF detection period are set based on the short cycle TA of the separately excited commutation signal FC. The same applies to the PWM inhibition period.
In addition, the separately excited commutation circuit 6 sets the period of the separately excited commutation signal FC to the long period TB after the time TC2. Thereby, since the BEMF detection period is maintained for a long time, the zero-cross detection accuracy is maintained high.
Thereafter, the separately excited commutation control by the separately excited commutation signal FC of the long period TB is continued until the self-excited commutation signal SC is generated again.

As shown in FIG. 26, when the detection of zero-crossing is not repeated, thereby failing in switching from separately excited commutation control to self-excited commutation control, the sensorless motor driving apparatus according to Embodiment 3 of the present invention Especially, the separately excited commutation control by the separately excited commutation signal FC of the long period TB is promptly restarted. That is, unlike the retry in the conventional sensorless motor driving device, for example, it is not necessary to determine the rotational speed of the rotor or to wait for a retry command from an external microprocessor. In this way, the sensorless motor can be started more smoothly.
In addition to the above, the sensorless motor driving apparatus according to Embodiment 3 of the present invention detects, for example, the rotational speed of the rotor, and based on the detected rotational speed, the other-excited commutation signal in the re-excited commutation control at the restart You may judge the cycle of FC.

  As described above, the sensorless motor driving apparatus and driving method according to the present invention achieve both suppression of motor echo noise and quick and reliable start-up of the sensorless motor by setting the PWM prohibition period and the BEMF detection period. Thus, the present invention is clearly industrially applicable.

It is a block diagram which shows the sensorless motor drive device by Embodiment 1 of this invention. 1 is a block diagram showing an internal configuration of a torque command circuit 12 in a sensorless motor driving apparatus according to Embodiment 1 of the present invention. FIG. 6 is a waveform diagram of a position signal PS, an original torque command TQ, an increase torque command TQ1, and a decrease torque command TQ2 for the sensorless motor driving device according to the first embodiment of the present invention. It is a table | surface which shows the state of each of phase current Iu, Iv, and Iw in each energized phase I-VI about the sensorless motor drive device by Embodiment 1 of this invention. It is a waveform diagram of phase currents Iu, Iv, Iw, and BEMFVu, Vv, Vw for the sensorless motor driving apparatus according to Embodiment 1 of the present invention. FIG. 5 shows an enlarged waveform diagram of phase currents Iu, Iv, Iw, and BEMFVu, Vv, Vw in the energized phases I to IV shown in FIG. FIG. 6 is a waveform chart of the BEMF detection signal DZC. FIG. 5 is a waveform diagram showing a case where a zero cross in the W phase occurs during the BEMF detection period, that is, a case of zero cross edge detection, in the sensorless motor driving apparatus according to the first embodiment of the present invention. FIG. 7 is a waveform diagram showing a case where a zero cross in the W phase occurs before the BEMF detection period, that is, a case of a zero cross state detection in the sensorless motor driving apparatus according to the first embodiment of the present invention. 3 is a block diagram showing an internal configuration of a separately excited commutation circuit 6 in the sensorless motor driving apparatus according to Embodiment 1 of the present invention. FIG. FIG. 5 is a timing diagram showing a separately excited commutation signal FC, a count CN1 for pulses of the clock signal CLK, and a self-excited commutation signal SC for the sensorless motor driving apparatus according to the first embodiment of the present invention. 6 is a first half of a flowchart showing an operation of generating a separately excited commutation signal FC by the separately excited commutation signal generation circuit 62 in the sensorless motor driving apparatus according to the first embodiment of the present invention. FIG. 10B is the latter half of the flowchart showing the operation of generating the separately excited commutation signal FC by the separately excited commutation signal generation circuit 62 in the sensorless motor driving apparatus according to Embodiment 1 of the present invention. 3 is a block diagram showing an internal configuration of a count circuit 72 in the sensorless motor driving apparatus according to Embodiment 1 of the present invention. FIG. Timing chart of commutation signal CS, count signal 72 internal signal CN2, RTC, DV, and three signals PS, NPWM, DZC sent from count circuit 72 for the sensorless motor driving apparatus according to Embodiment 1 of the present invention It is. FIG. 4 is a waveform diagram of phase current and BEMF for the same motor coil when the rotor of the sensorless motor M stably rotates in the sensorless motor driving apparatus according to Embodiment 1 of the present invention. About the sensorless motor drive device by Embodiment 1 of this invention, when the sensorless motor M starts, the separately excited commutation signal FC, the self-excited commutation signal SC, the commutation signal CS, and the phase currents Iu, Iv, and Iw are shown. It is a timing chart. In particular, the case is shown where the pulse interval T2 of the self-excited commutation signal SC is shorter than the original pulse interval T of the separately excited commutation signal FC (T2 <T). About the sensorless motor drive device by Embodiment 1 of this invention, when the sensorless motor M starts, the separately excited commutation signal FC, the self-excited commutation signal SC, the commutation signal CS, and the phase currents Iu, Iv, and Iw are shown. It is a timing chart. In particular, when the pulse interval T2 of the self-excited commutation signal SC is longer than the original pulse interval T of the separately excited commutation signal FC and shorter than the extended pulse interval 6T of the other excitation commutation signal FC (T <T2 < 6T). About the sensorless motor drive device by Embodiment 1 of this invention, when the sensorless motor M starts, the separately excited commutation signal FC, the self-excited commutation signal SC, the commutation signal CS, and the phase currents Iu, Iv, and Iw are shown. It is a timing chart. In particular, a case where a new self-excited commutation signal SC is not generated from time TC until the elapse of the extended period 6T of the separately-excited commutation signal FC is shown. 7 is a block diagram showing an internal configuration of a count circuit 72 in a sensorless motor driving apparatus according to Embodiment 2 of the present invention. FIG. 6 is a timing chart of a commutation signal CS, a 4-bit parallel signal DV, a position signal PS, a PWM inhibition signal NPWM, a trigger signal TR, and a BEMF detection signal DZC for a sensorless motor driving apparatus according to Embodiment 2 of the present invention. 6 is a block diagram showing an internal configuration of a separately excited commutation circuit 6 in a sensorless motor driving apparatus according to Embodiment 3 of the present invention. FIG. In the sensorless motor driving apparatus according to the third embodiment of the present invention, two signals PA and PB generated by the separately excited commutation signal generation circuit 62A, a first count CN1A and a second count CN1B for the pulse of the clock signal CLK FIG. 5 is a timing chart showing a separately excited commutation signal FC and a self-excited commutation signal SC. In the first part of the flowchart showing the generation operation of the separately excited commutation signal FC by the separately excited commutation signal generation circuit 62A and the separately excited commutation signal control circuit 63 in the sensorless motor driving apparatus according to Embodiment 3 of the present invention. is there. In the second part of the flowchart showing the generation operation of the separately excited commutation signal FC by the separately excited commutation signal generation circuit 62A and the separately excited commutation signal control circuit 63, the sensorless motor driving apparatus according to Embodiment 3 of the present invention is described. is there. In the third part of the flowchart showing the generation operation of the separately excited commutation signal FC by the separately excited commutation signal generation circuit 62A and the separately excited commutation signal control circuit 63, the sensorless motor driving apparatus according to Embodiment 3 of the present invention is described. is there. In the sensorless motor driving apparatus according to the third embodiment of the present invention, the timing indicating the separately excited commutation signal FC, the self-excited commutation signal SC, the commutation signal CS, and the phase currents Iu, Iv, and Iw when the sensorless motor is activated. It is a chart. It is a block diagram which shows the conventional sensorless motor drive device. For the conventional sensorless motor driving device, the BEMFVu, Vv, Vw and currents Iu, Iv, Iw, position detection mask signal MZC, and three of the three motor coils Mu, Mv, Mw at the time of stable rotation of the rotor FIG. 6 is a waveform diagram showing potentials VU0, VV0, and VW0 of drive terminals U0, V0, and W0, respectively.

Explanation of symbols

SP1 First set pulse signal
SP2 Second set pulse signal
TQ Original torque command
TQ1 Increase torque command
TQ2 Decrease torque command
13 Current comparator
IC1 First output signal of current comparator 13
IC2 Second output signal of current comparator 13
IC3 Third output signal of current comparator 13
P1 First PWM control signal
P2 Second PWM control signal
CP Energized phase switching signal
3 Output circuit
33 Power supply terminal
31U U-phase high-side power transistor
31V V-phase high-side power transistor
31W W-phase high-side power transistor
32U U-phase low-side power transistor
32V V-phase low-side power transistor
32W W-phase low-side power transistor
R Current detection resistor
M Sensorless motor
Mu U phase motor coil
Mv V phase motor coil
Mw W phase motor coil
U0 U-phase drive terminal
V0 V-phase drive pin
W0 W phase drive pin
C Motor coil mid-point terminal
Iu U phase current
Iv V phase current
Iw W phase current
4 BEMF comparator
First output signal of BCU BEMF comparator 13
Second output signal of BCV BEMF comparator 13
BCW BEMF comparator 13 third output signal
ZCP zero cross point information
MPWM PWM mask signal
DZC BEMF detection signal
SC self-excited commutation signal
FC Other commutation signal
CS commutation signal
PS position signal
NPWM PWM inhibition signal

Claims (28)

An output circuit for energizing the motor coil of the sensorless motor;
A PWM control unit that generates a PWM control signal indicating a timing of energization of the motor coil based on a position signal indicating an estimated position of the rotor of the sensorless motor;
Energized phase switching circuit that switches the energized phase in synchronization with the commutation signal;
The motor coil corresponding to the energized phase is selected, and the energization of the selected motor coil is changed by the output circuit according to the PWM control signal, and in particular during the PWM prohibition period, the specific output of the specific motor coil by the output circuit is changed. Pre-drive circuit that inhibits energization;
A BEMF comparator that detects BEMF (back electromotive force) induced in the motor coil and compares the BEMF with a midpoint voltage of the motor coil;
A self-excited commutation circuit that detects coincidence of the BEMF and the midpoint voltage during a BEMF detection period, that is, detects a zero cross, and generates a self-excited commutation signal when the zero cross is detected; and
The commutation signal is generated based on the self-excited commutation signal, the interval of the commutation signal is measured, the position signal is generated based on the interval, and the PWM inhibition period and the BEMF detection are detected based on the position signal. A counting unit that sets a period, in particular, starts the PWM prohibition period earlier than the BEMF detection period, and ends both the PWM prohibition period and the BEMF detection period in synchronization with the commutation signal;
A sensorless motor driving device having A command circuit for setting a target current based on the original command and the position signal;
A current comparator that detects a current of the motor coil and compares the detected current with the target current; and a PWM control circuit that generates the PWM control signal according to a difference between the detected current and the target current;
The sensorless motor driving device according to claim 1, wherein the PWM control unit includes:   The sensorless motor driving device according to claim 2, wherein the command circuit gradually increases or decreases the target current.   When the predetermined time elapses from the start time of the PWM prohibition period or when the estimated position of the rotor changes by a predetermined amount from the value at the start time of the PWM prohibition period, whichever is earlier, The sensorless motor driving device according to claim 1, wherein the BEMF detection period is started. An output circuit for energizing the motor coil of the sensorless motor;
A PWM control unit that generates a PWM control signal indicating a timing of energization of the motor coil based on a position signal indicating an estimated position of the rotor of the sensorless motor;
Energized phase switching circuit that switches the energized phase in synchronization with the commutation signal;
A pre-drive circuit that selects the motor coil corresponding to the energized phase and changes the energization of the selected motor coil by the output circuit according to the PWM control signal;
A BEMF comparison unit that detects BEMF induced in the motor coil and compares the BEMF with a midpoint voltage of the motor coil;
A self-commutated commutation circuit that detects coincidence of the BEMF and the midpoint voltage during a BEMF detection period, that is, detects a zero cross and generates a self-excited commutation signal when the zero cross is detected;
A separately excited commutation circuit for generating a separately excited commutation signal at a predetermined period;
A selection circuit that selects, as the commutation signal, one of the self-commutated commutation signal and the other excitation commutation signal that is input first during the BEMF detection period;
The interval of the commutation signal is measured, the position signal is generated based on the interval, the BEMF detection period is set based on the position signal, and in particular, the BEMF detection period is terminated in synchronization with the commutation signal. Counting circuit;
A sensorless motor driving device. A command circuit for setting a target current based on the original command and the position signal;
A current comparator that detects a current of the motor coil and compares the detected current with the target current; and a PWM control circuit that generates the PWM control signal according to a difference between the detected current and the target current;
The sensorless motor driving device according to claim 5, wherein the PWM control unit includes:   The sensorless motor driving device according to claim 6, wherein the command circuit gradually increases or decreases the target current. The count circuit sets a PWM prohibition period based on the position signal, in particular, starts the PWM prohibition period earlier than the BEMF detection period, and ends the PWM prohibition period in synchronization with the commutation signal;
The pre-drive circuit prohibits energization of the specific motor coil by the output circuit during the PWM inhibition period;
The sensorless motor driving device according to claim 5.   The count circuit at a point in time when a certain time has elapsed since the start of the PWM prohibition period or when the estimated position of the rotor has changed by a certain amount from a value at the start of the PWM prohibition period. The sensorless motor driving device according to claim 8, wherein the BEMF detection period is started.   The sensorless motor driving device according to claim 5, wherein the separately excited commutation circuit constantly generates the separately excited commutation signal.   When the self-excited commutation signal is input prior to the other-excited commutation signal during the BEMF detection period, the separately-excited commutation circuit extends the period of the other-excited commutation signal in the next BEMF detection period. The sensorless motor driving device according to claim 5.   When the other-excited commutation signal is input a predetermined number of times before the self-excited commutation signal during the BEMF detection period, the other-excited commutation circuit causes the other excitation commutation circuit in the next BEMF detection period. The sensorless motor driving device according to claim 5, wherein the cycle of the flow signal is extended.   The sensorless motor driving device according to claim 5, wherein the separately-excited commutation circuit generates at least two types of pulse signals having different periods and selects one of the pulse signals as the separately-excited commutation signal.   The sensorless motor driving device according to claim 13, wherein the pulse signal includes a first signal having a constant period and a second signal having a period twice as long as the period of the first signal. Generating a PWM control signal indicating timing of energization of the motor coil of the sensorless motor based on a position signal indicating the estimated position of the rotor of the sensorless motor;
Selecting the motor coil corresponding to the energized phase and energizing the selected motor coil in accordance with the PWM control signal;
Starting a PWM inhibition period based on the position signal;
Prohibiting energization of a specific motor coil during the PWM inhibition period;
Detecting BEMF induced in the motor coil and comparing the BEMF with a midpoint voltage of the motor coil;
Step to initiate the B EMF detection period later than the start of the PWM prohibition period;
The step of the coincidence of the BEMF during said detection period BEMF said midpoint voltage and, that detects the zero crossing, to generate a self励転flow signal upon detection of the zero crossing;
Generating a commutation signal based on the self-excited commutation signal;
Ending both the PWM inhibition period and the BEMF detection period in synchronization with the commutation signal;
Switching the energized phase in synchronization with the commutation signal;
Measuring the interval of the commutation signal; and
Generating the position signal based on an interval of the commutation signal;
A sensorless motor driving method. A sub-step for setting a target current based on the original command and the position signal;
Detecting a current of the motor coil and comparing the detected current with the target current; and
A sub-step of generating the PWM control signal according to a difference between the detected current and the target current;
16. The sensorless motor driving method according to claim 15, wherein the step of generating the PWM control signal includes:   The sensorless motor driving method according to claim 16, wherein the target current is gradually increased or decreased for each sub-step for setting the target current. The sensorless motor driving method includes a step of measuring an elapsed time and a change amount of the estimated position of the rotor from a start time of the PWM inhibition period;
The BEMF detection is performed at the earlier of a certain time after the start of the PWM inhibition period or a time when the estimated position of the rotor changes by a certain amount from the value at the start of the PWM inhibition period. The step of starting the period is performed;
The sensorless motor driving method according to claim 15. Generating a PWM control signal indicating timing of energization of the motor coil of the sensorless motor based on a position signal indicating the estimated position of the rotor of the sensorless motor;
Selecting the motor coil corresponding to the energized phase and energizing the selected motor coil in accordance with the PWM control signal;
Detecting BEMF induced in the motor coil and comparing the BEMF with a midpoint voltage of the motor coil;
Starting a BEMF detection period based on the position signal;
Detecting coincidence of the BEMF and the midpoint voltage during the BEMF detection period, that is, a zero cross, and generating a self-excited commutation signal when the zero cross is detected;
Generating a separately excited commutation signal at a predetermined period;
Selecting one of the self-excited commutation signal and the other-excited commutation signal that is generated earlier during the BEMF detection period as a commutation signal;
Ending the BEMF detection period in synchronization with the commutation signal;
Switching the energized phase in synchronization with the commutation signal;
Measuring the interval of the commutation signal; and
Generating the position signal based on an interval of the commutation signal;
A sensorless motor driving method. A sub-step for setting a target current based on the original command and the position signal;
Detecting a current of the motor coil and comparing the detected current with the target current; and
A sub-step of generating the PWM control signal according to a difference between the detected current and the target current;
20. The sensorless motor driving method according to claim 19, wherein the step of generating the PWM control signal includes:   21. The sensorless motor driving method according to claim 20, wherein the target current is gradually increased or decreased for each sub-step for setting the target current. Starting a PWM inhibition period earlier than the BEMF detection period based on the position signal;
Prohibiting energization of a particular motor coil during the PWM inhibition period; and
Ending the PWM inhibition period in synchronization with the commutation signal;
The sensorless motor driving method according to claim 19, comprising: The sensorless motor driving method includes a step of measuring an elapsed time and a change amount of the estimated position of the rotor from a start time of the PWM inhibition period;
The BEMF detection is performed at the earlier of a certain time after the start of the PWM inhibition period or a time when the estimated position of the rotor changes by a certain amount from the value at the start of the PWM inhibition period. The step of starting the period is performed;
The sensorless motor driving method according to claim 22.   The sensorless motor driving method according to claim 19, wherein the step of generating the separately excited commutation signal is constantly executed.   When the self-excited commutation signal is generated prior to the other-excited commutation signal during the BEMF detection period, the period of the separately-excited commutation signal is extended in the next BEMF detection period. Item 20. A sensorless motor driving method according to Item 19.   When the separately-excited commutation signal is generated a predetermined number of times before the self-excited commutation signal during the BEMF detection period, the period of the separately-excited commutation signal is extended in the next BEMF detection period. The sensorless motor driving method according to claim 19, further comprising: a step.   The step of generating the separately excited commutation signal includes a substep of generating at least two types of pulse signals having different periods, and a substep of selecting any one of the pulse signals as the separately excited commutation signal. 20. A sensorless motor driving method according to claim 19, further comprising:   28. The sensorless motor driving method according to claim 27, wherein the pulse signal includes a first signal having a constant period and a second signal having a period twice as long as the period of the first signal.

Images