JP4927000B2 - Sensorless brushless motor control circuit, sensorless brushless motor device, and vacuum pump device - Google Patents

Description

  The present invention relates to a brushless motor, a control circuit thereof, and the like, for example, to a vacuum pump such as a magnetic bearing type turbo molecular pump or a magnetic bearing spindle.

Conventionally, a brushless motor is started as follows.
Some brushless motors include a rotor having a two-pole permanent magnet and three motor windings that generate a magnetic field for rotating the rotor around the rotor.
Among such brushless motors, as a sensorless brushless motor control circuit without a sensor for detecting the position of the magnetic pole, a motor driving current is passed through two motor windings of three motor windings to rotate the rotor, For example, the rotor magnetic pole position is detected from the induced electromotive force generated in the remaining one motor winding due to the rotation of the rotor, and the current of the motor winding is sequentially switched based on this magnetic pole position.

An example of the conventional brushless motor control circuit described above will be described with reference to FIGS.
FIG. 8 is a conceptual diagram showing a three-phase full-wave brushless motor. The rotor 150 includes a two-pole permanent magnet. Around the rotor, U-phase, V-phase, and W-phase motor windings 151U, 151V, 151W are arranged. Two motor windings are excited by flowing current, and the rotor 150 is rotated by the attractive force of the magnetic force. The excited motor windings 151U, 151V, 151W are sequentially switched according to the position of the magnetic poles of the rotor 150, and the rotation of the rotor 150 is continued. The position of the magnetic pole is detected by detecting the voltage induced in the remaining one motor winding that is not excited.

As shown in FIG. 9, there are six types of drive voltage vectors output to the motor windings 151U, 151V, 151W of the three-phase full-wave brushless motor.
The drive voltage vector is the drive voltage vector when current is passed from the U phase motor winding to the V phase motor winding, and the drive voltage vector is the drive voltage vector when current is passed from the U phase motor winding to the W phase motor winding. 2. Driving voltage vector when driving current from V-phase motor winding to W-phase motor winding, driving voltage vector 3, driving voltage vector when driving current from V-phase motor winding to U-phase motor winding Voltage vector 4, drive voltage vector when current flows from W-phase motor winding to U-phase motor winding, drive voltage vector 5, drive voltage vector when current flows from W-phase motor winding to V-phase motor winding Is a drive voltage vector 6, and hereinafter, each drive voltage vector is distinguished by a number.
These drive voltage vector numbers are circled in FIG.
Further, the current that flows from the V-phase motor winding to the W-phase motor winding is referred to as a current in the V → W direction.

The motor control circuit synchronizes with the rotation of the rotor 150 from the detected magnetic pole position, and generates one pulse for each rotation of the rotor 150. This pulse is input to a PLL (Phase Lock Loop) circuit (not shown), and the PLL circuit generates six pulses having a period six times the rotation of the rotor 150. The above six driving voltage vectors are sequentially switched in synchronism with these six pulses, and the rotation of the rotor 150 is continued. That is, the magnetic pole position of the rotor 150 is detected from the voltage of the motor winding in the non-energized phase, and the voltage vector output to the motor windings 151U, 151V, 151W is switched while applying feedback based on the detected value.
By the way, in order to lock (activate) this PLL circuit, the frequency of the input signal needs to be at least about 20 Hertz. That is, the PLL circuit cannot be operated unless the rotor 150 rotates about 20 times per second.
Conventionally, each drive voltage vector is switched in an open loop until the motor is started and the rotational speed of the rotor 150 increases to a rotational speed at which the PLL circuit can be locked. That is, without applying any feedback, the voltage vector applied to the motor windings 151U, 151V, 151W is first switched sequentially at a low speed close to DC (direct current), and the rotor is sucked into this while gradually increasing the switching speed. Following this, acceleration was performed to a rotational speed at which the PLL circuit can be locked.

Japanese Patent Laid-Open No. 8-47285 discloses a control circuit for a brushless motor that uses a PLL circuit to generate a synchronous pulse of the rotor rotation speed and switch the drive voltage vector. In this method, the position of the magnetic pole is detected by a Hall sensor, and the drive voltage vector is feedback-controlled.
Three hall sensors are arranged around the magnetic pole of the rotor at an interval of 120 °. When the rotor is rotating at a low speed at which the PLL circuit cannot be locked, the detection by these three hall sensors is performed. When the drive voltage vector is controlled by the signal and the rotor speed reaches a speed at which the PLL circuit can be locked, the PLL circuit has a period three times the rotor speed from the detection signal of one of the hall sensors. A multiplied synchronization pulse is generated, and the drive voltage vector is switched by this multiplied synchronization pulse.
This technique can also be applied by detecting a counter electromotive voltage generated by rotation of a rotor generated in a motor winding without using a hall sensor. That is, the present invention can be applied to a hall sensorless motor drive circuit using a PLL circuit.

Conventional sensorless brushless motors have been controlled by a control circuit that operates as follows.
The sensorless brushless motor control circuit feedback controls the current flowing through the motor winding while detecting the magnetic pole position of the rotor. The magnetic pole position of the rotor is detected by detecting a voltage induced in the motor winding by the rotation of the rotor, that is, an induced electromotive force. For example, in the case of a three-phase brushless motor, a voltage is applied to two motor windings, and a voltage induced in the remaining non-conducting phase is detected. Then, based on the magnetic pole position detected by this voltage, the two motor windings to which the voltage is to be applied next are determined, and the voltage is applied. At that time, the induced electromotive force of the motor winding in the non-energized phase is detected, thereby detecting the position of the magnetic pole. The motor is driven by performing this process continuously.
FIG. 20 is a diagram showing the magnetic pole position detection timing of the control circuit of the conventional sensorless brushless motor. Waveforms 201a, 201b, and 201c are waveform diagrams of voltages induced in a certain motor winding. As will be described later, FIG. 20A shows the case where the phase of the rotating magnetic field due to the motor winding current is ahead of the phase of the rotor rotation, and FIG. 20B shows the case where both phases are aligned. FIG. 20C shows a case where the phase of the rotating magnetic field is ahead of the phase of the rotor.
The magnetic pole position is detected by sampling the intersections 203a, 203b, 203c of the virtual neutral potentials 202a, 202b, 202c and the waveforms 201a, 201b, 201c.
This control circuit has a drive mode for outputting a voltage to the motor winding and a sampling mode for not outputting the voltage. As shown in FIG. 20, the voltage is output to the motor winding in the driving mode during 2/3 of the rotor rotation period, and the voltage is not output in the sampling mode during the remaining 1/3. This is to prevent noise from being superimposed on the waveforms 201a, 201b, and 201c when the magnetic pole position is detected.
The detection of the intersections 203a, b, c is performed in the sampling mode.
FIG. 20A shows a case where the phase of the rotating magnetic field is ahead of the phase of rotation of the rotor. And the area surrounded by the virtual neutral potential 202a. FIG. 20B shows a case in which the phases of the two coincide with each other, and the left and right areas are equal. FIG. 20C shows a case where the phase of the rotating magnetic field is delayed from the phase of the rotor, and the area on the left side of the intersection 203c is larger than the area on the right side.
The conventional control circuit feedback-controls the voltage output to the motor winding so that the areas on the left and right of the intersection are always equal like the intersection 203b.

Furthermore, when the conventional sensorless brushless motor is applied to a vacuum pump such as a turbo molecular pump, there are the following problems.
The motor unit of the turbo molecular pump may be constituted by a DC brushless motor composed of a rotor shaft having a permanent magnet and a plurality of electromagnets arranged around the permanent magnet with a predetermined gap therebetween. is there.

  However, in the conventional starting method, when the switching speed of the drive voltage vectors of the motor windings 151U, 151V, 151W is suddenly increased or the load of the rotor 150 is suddenly changed, the rotor 150 is rotated by the motor winding 151U. , 151V, 151W could not follow the magnetic field produced, step out, sometimes failed to start. Further, when the voltage vector switching speed is gradually increased over a long time, it takes a long time for the rotor 150 to reach a rotational speed at which the PLL circuit can be locked. Furthermore, when a power failure occurs after the motor is started and before the rotor 150 reaches the number of revolutions that can lock the PLL circuit, it is necessary to restart the sensor 150. The conventional sensorless brushless motor control circuit cannot detect the magnetic pole position. Therefore, it is necessary to restart the rotor 150 after stopping it by DC braking once. In particular, in the case of a turbo molecular pump, it took about 1 minute to accelerate the rotation speed of the rotor 150 to about 20 rotations per second at which the PLL circuit can be locked.

  On the other hand, in the control circuit of the conventional sensorless brushless motor, the intersections 203a, 203b, and 203c must be in the sampling mode. For example, when a load fluctuation occurs in the rotor and the intersections 203a, 203b, and 203c are out of the sampling mode, There was a case where the position was disturbed and a step-out state occurred. In addition, when the magnetic pole of the rotor is detected, noise is superimposed on the voltage of the motor winding, and the magnetic pole position may not be detected accurately.

  In addition, when the brushless motor rotor is pivotally supported by a magnetic bearing, for example, if the rotor is DC braked at start-up and the magnetic pole is set at a predetermined position, the magnetic bearing has no friction, so the rotor is positioned at a predetermined position. There was a problem that it vibrated as the center and the vibration was not damped quickly. Also, the magnetic field is slowly rotated in an open loop until the rotational speed of the rotor shaft reaches the rotational frequency at which the PLL circuit can be locked (the rotational speed of the rotor per unit time, here about 20 [Hz]). Therefore, it takes time to start up, and if the rotational speed of the rotor shaft changes greatly during steady operation, the position of the magnetic pole cannot be detected, and the step-out may occur.

  Therefore, the first object of the present invention is to accurately detect the magnetic pole position of the rotor even when the rotational speed of the rotor changes greatly due to load fluctuations or noise is superimposed on the voltage of the motor winding. Then, it is providing the control apparatus of a sensorless brushless motor which can control the electric current of a motor winding appropriately.

  The second object of the present invention is to detect the magnetic pole of the rotor and perform feedback control of the switching of the magnetic field even at a low speed rotation of 20 revolutions per second or more. A sensorless brushless motor control circuit, a sensorless brushless motor device, and a vacuum pump device using the motor can accurately detect the position of the magnetic pole and perform feedback control even when noise is superimposed on the voltage of It is.

The present invention achieves the first object, a rotor provided with magnetic poles, a plurality of motor windings for rotating the rotor, current supply means for supplying a current to the plurality of motor windings, Magnetic flux acquisition means for acquiring the interlinkage magnetic flux of at least one of the motor windings by the magnetic pole, and magnetic pole position acquisition means for acquiring the position of the magnetic pole from the change in the interlinkage magnetic flux acquired by the magnetic flux acquisition means And the magnetic flux acquisition means includes: voltage processing calculation means for performing voltage processing calculation for subtracting a voltage based on a voltage drop due to resistance at least from the line voltage of the motor winding; and when performing the voltage processing calculation in, and a DC blocking unit for removing a DC component superimposed from a signal for calculating the flux linkage, the current supply means, it has been the magnetic poles of the position acquired by the magnetic pole position acquiring means Based on, to provide a control circuit of the sensorless brushless motor and switches the current of the motor windings (first configuration).
As a modification of the first configuration, the magnetic flux acquisition means may acquire the difference between the flux linkages of the two motor windings by the magnetic poles.
In the control circuit of the sensorless brushless motor of the present invention, the rotational position of the magnetic poles of the rotor can be obtained at any time during motor operation. The current of the motor winding can be controlled appropriately.

Further, the magnetic flux acquiring means in a first configuration comprises a first acquisition means for acquiring a line voltage value of a predetermined two of the motor windings, the resistance of the two predetermined motor windings, the current supply A second obtaining means for obtaining a voltage drop due to a combined resistance of the power supply device constituting the means and the resistance of the cable connecting the motor winding; and a difference between the currents of the two predetermined motor windings, Third acquisition means for acquiring a value obtained by multiplying the inductance values of the two motor windings, and subtracting the value acquired by the second acquisition means from the value acquired by the first acquisition means After that, an integration value acquisition unit that integrates and a subtraction unit that subtracts the value acquired by the third acquisition unit from the value acquired by the integration value acquisition unit (second configuration) ).

As a modification of the second configuration, the magnetic flux acquisition unit in the first configuration includes a first integrated value acquisition unit that acquires a value obtained by integrating the line voltages of the two predetermined motor windings with time, A value obtained by integrating the voltage drop due to the combined resistance of the resistance of the two predetermined motor windings, the power supply device constituting the current supply means and the resistance of the cable connecting the motor winding with time is obtained. Integration value acquisition means, third integration value acquisition means for acquiring a value obtained by integrating the voltage drop due to inductance of the two predetermined motor windings with time, and values acquired by the first integration value acquisition means From the above, the value acquired by the second integral value acquisition unit and the value acquired by the third integral value acquisition unit may be subtracted.
In the second configuration and the modification of the second configuration, when the integral value is acquired, the signal is integrated using an integrator, so that the noise superimposed on the signal is canceled and the signal with less noise is obtained. Can be obtained. For this reason, it is possible to monitor the rotation of the rotor while operating the motor.

In addition, the value of the combined resistance used in the second configuration is obtained by dividing the direct current supply means for supplying a direct current to two predetermined motor windings and the line voltage value by the current value of the direct current. Thus, it can be obtained by a combined resistance value acquiring means including a first calculating means for calculating the value of the combined resistance.
This is done, for example, by passing a direct current through two predetermined motor windings before starting the motor.

Further, the inductance used in the second configuration includes high-frequency current supply means for supplying a high-frequency current to two predetermined motor windings, and a line voltage between the two motor windings when the high-frequency current is supplied. A line voltage value acquisition means for acquiring a value; a second calculation means for acquiring a value obtained by dividing the line voltage value by the current value of the high-frequency current and the frequency of the high-frequency current and a predetermined constant; It can be acquired by the provided inductance acquisition means.
This method can be performed, for example, by conducting a high-frequency current that cannot be followed by the rotor before starting the motor to two predetermined motor windings. The predetermined value is 2π.

In addition, the inductance used in the second configuration includes the rotor rotation means that rotates the rotor by switching the current of the motor winding in an open loop, and the integral value acquisition before and after switching the current of the motor winding. Sampling means for sampling the integral value obtained by the means, current peak value obtaining means for obtaining the peak value of the current value supplied to the two predetermined motor windings, and the current obtained by the sampling means The absolute value of the difference between the integrated values before and after switching can be obtained by an inductance obtaining means comprising: a third computing means for dividing the absolute value by the current peak value obtained by the current peak value obtaining means.
In this method, the rotor is rotated in an open loop up to a certain number of rotations, and the inductance is obtained based on the size of the step appearing in the waveform obtained by the first integration means when the motor drive current is switched.

The combined resistance value and the inductance used in the second configuration are assumed magnetic flux acquisition means for acquiring a linkage flux between two predetermined motor windings using the assumed value of the combined resistance and the assumed value of the inductance. And when the rotor rotating means rotates the rotor at a predetermined angular velocity, the line voltage value of the predetermined two motor windings and the current supply of the motor winding are stopped to Line voltage between two predetermined motor windings when free running at a predetermined angular velocity, a signal obtained by the assumed magnetic flux acquisition means when supply of the current is stopped, and supply of the current can be obtained by providing a correction means for correcting the phase difference of the signal obtained at the assumed magnetic flux acquiring means when resumed, the assumed value of the assumed value and the inductance of the combined resistance from the
First, the interlinkage magnetic flux generated in the motor winding is obtained using the assumed combined resistance value and inductance, and thereby the assumed combined resistance value and inductance are corrected. By repeating this process several times, the combined resistance value and the inductance that have been sequentially corrected approach the true value.

According to the present invention, in order to achieve the second object, among the plurality of motor windings for rotating the rotor provided with the magnetic poles, the phase and magnitude of the voltage drop due to the inductance of the motor winding are a voltage difference equal predetermined two phases, and subtraction of the voltage based on the voltage drop due to at least resistance, and the arithmetic processing consisting of integration, and the magnetic flux signal acquiring means for acquiring a magnetic flux signal, when performing the arithmetic processing to a DC blocking unit for removing a DC component is superimposed on the signal of the voltage difference from the arithmetic processing magnetic flux signals on the basis of the removed DC component superimposed signal by the direct current cutoff section can output driving voltage First drive timing acquisition means for acquiring drive timings of some of the drive voltage vectors of the vectors, and drive timing acquired by the first drive timing acquisition means The first drive voltage vector output means for outputting the part of the drive voltage vectors in synchronization with the output voltage, and the drive voltage that can be output by multiplying the timing obtained from the magnetic flux signal acquired by the magnetic flux signal acquisition means A second drive timing acquisition means for acquiring a drive timing of the vector, and a second drive voltage vector for outputting the output drive voltage vector in synchronization with the drive timing acquired by the second drive timing acquisition means Provided is a sensorless brushless motor control circuit comprising output means, selection means for selecting the first drive voltage vector output means, and the second drive voltage vector output means ( Third configuration).

  Further, in order to achieve the second object, the present invention provides a current supply means for supplying a current to a plurality of motor windings for rotating a rotor having magnetic poles, and among the plurality of motor windings, Line voltage acquisition means for acquiring the line voltage of two predetermined motor windings having the same phase and magnitude of the voltage drop due to the inductance of the motor winding, and the line width acquired by the line voltage acquisition means Resistance correction for correcting a change in voltage due to a combined resistance of a resistance of the two predetermined motor windings, a power supply device constituting the current supply means, and a resistance of a connection cable connecting the motor windings from the voltage Out of the magnetic flux signal acquisition means for acquiring the magnetic flux signal by integrating the line voltage corrected by the resistance component correction means and the magnetic flux signal acquired by the magnetic flux signal acquisition means. Mo Reactance correction means for correcting a change due to the reactance of the winding, magnetic pole position acquisition means for acquiring the position of the magnetic pole from the magnetic flux signal corrected by the reactance correction means, the resistance correction means, A correction invalidity means for disabling at least the reactance correction means and not performing correction among the reactance correction means, and when the rotation speed of the rotor is equal to or less than a predetermined rotation, the correction is invalid Means for disabling at least the reactance correction means, and the current supply means switches a current flowing through the predetermined two motor windings based on the magnetic pole position acquired by the magnetic pole position acquisition means. When the current is supplied to the two motor windings in the mode and the rotation speed of the rotor is larger than a predetermined rotation, the correction invalidation means is used. And supplying a current to the motor winding in a second mode in which the current of the motor winding is switched based on the magnetic pole position acquired by the magnetic pole position acquisition means. A circuit is provided (fourth configuration).

  In the fourth configuration, when the mode is switched from the first mode to the second mode, the current supply means causes a minute current to flow through the plurality of motor windings in a predetermined order for a predetermined time. (5th structure).

  In the fourth configuration or the fifth configuration, the first cut-off frequency and the second cut-off frequency higher than the first cut-off frequency can be switched, and the direct current component superimposed on the magnetic flux signal And a switching means for switching between the first cutoff frequency and the second cutoff frequency of the DC cutoff means (sixth configuration). .

  In the sixth configuration, the switching means sets the cut-off frequency of the DC cut-off means as the first cut-off frequency for a predetermined time after the rotor starts, and when the predetermined time elapses, It can be configured to switch to a cut-off frequency.

In order to achieve the second object, the present invention provides a motor unit comprising a rotor having magnetic poles and a plurality of motor windings for rotating the rotor, and the plurality of motor windings. Current supply means for supplying current to the magnetic flux, magnetic flux acquisition means for acquiring the interlinkage magnetic flux of at least one of the motor windings by the magnetic pole, and the change in the interlinkage magnetic flux acquired by the magnetic flux acquisition means A magnetic pole position acquisition means for acquiring the position of the magnetic pole, and the magnetic flux acquisition means performs a voltage processing operation for subtracting at least a voltage based on a voltage drop due to a resistance from a line voltage of the motor winding. a voltage processing operation means for performing, when performing said voltage processing operation, and a DC blocking unit for removing a DC component superimposed from a signal for calculating the flux linkage, the current supply means Based on the position of the magnetic pole acquired by the magnetic pole position acquiring means, to provide a sensorless brushless motor apparatus characterized by switching a current of the motor windings (seventh configuration).

In order to achieve the second object, the present invention provides a rotor having a magnetic pole, a plurality of motor windings for rotating the rotor, and the motor winding among the plurality of motor windings. The voltage difference between two predetermined phases that is equal to the phase and magnitude of the voltage drop due to the inductance of the wire is at least calculated by subtracting the voltage based on the voltage drop due to the resistance, and integrating , to obtain the magnetic flux signal Based on the magnetic flux signal acquisition means to be acquired, the DC blocking unit that removes the DC component superimposed on the voltage difference signal when performing the arithmetic processing, and the signal from which the DC component superimposed by the DC blocking unit is removed from the arithmetic processing magnetic flux signal, and a first drive timing acquiring means for acquiring a drive timing of a portion of the driving voltage vectors among outputable driving voltage vectors, drive the first First drive voltage vector output means for outputting the partial drive voltage vector in synchronization with the drive timing acquired by the timing acquisition means, and multiplying the timing obtained from the magnetic flux signal acquired by the magnetic flux signal acquisition means Second drive timing acquisition means for acquiring the drive timing of the drive voltage vector that can be output, and the drive voltage vector that can be output in synchronization with the drive timing acquired by the second drive timing acquisition means A second drive voltage vector output means for outputting the signal, a selection means for selecting the first drive voltage vector output means, and the second drive voltage vector output means. A brushless motor device is provided (eighth configuration).

  In order to achieve the second object, the present invention provides an exterior body in which an intake port is formed at one end and an exhaust port is formed at the other end, and a magnetic bearing or a mechanical bearing inside the exterior body. A rotor rotatably supported by the motor, a motor for rotating the rotor, and a stator disposed in the exterior body, wherein the motor has a seventh configuration or an eighth configuration. A vacuum pump device characterized by comprising a sensorless brushless motor device is provided.

The control circuit for the sensorless brushless motor according to any one of the first to sixth aspects of the invention constantly monitors the magnetic pole position of the rotor while the sensorless brushless motor is driven. Stable control can be achieved without stepping out even when the value changes greatly. Further, even if noise is superimposed on the motor winding, this is removed by integration, so that the magnetic pole position of the rotor can be accurately detected.
The sensorless brushless motor control circuit of the present invention automatically measures the combined resistance value Rp of the resistance value of the motor winding and the resistance value of the cable connecting the motor and the inductance Lp of the motor winding. Even if the cable length is changed at the site or the motor is replaced with another motor, it can be used immediately without measuring Rp and Lp again using a measuring instrument.
Further, according to the present invention, the startup speed is fast and the brushless motor can be operated stably.

(First embodiment)
A first embodiment of a brushless motor control circuit according to the present invention will be described below with reference to FIGS. FIG. 1 is a diagram showing a main configuration of a control circuit of the brushless motor according to the first embodiment.
The control circuit 141 according to the present embodiment includes a rotor 112 having a two-pole permanent magnet, a motor 105 including star-connected motor windings 107U, 107V, and 107W for rotating the rotor 112, and these motor windings. A motor drive circuit 115 that supplies current to 107U, 107V, and 107W, a microcomputer 130 that controls the motor drive circuit 115, and resistors 108U, 108V, and 108W that have the same resistance value and are star-connected, are provided.
In FIG. 1, the motor windings 107 </ b> U, 107 </ b> V, 107 </ b> W and the rotor 112 are separately shown for convenience, but the motor winding 7 is actually disposed on the outer periphery of the rotor 112.
The motor drive circuit 115 includes a DC power source 116 and six transistors 121a, 121b, 121c, 121d, 121e, and 121f that form a three-phase bridge. The bases of the transistors 121a, 121b, 121c, 121d, 121e, and 121f are connected to the microcomputer 130, respectively. Each transistor 121a, 121b, 121c, 121d, 121e, 121f is turned on / off by a gate drive pulse from the microcomputer 130, and supplies a predetermined current to the motor windings 107U, 107V, 107W.
The motor drive circuit 115 supplies a predetermined current to the motor windings 107U, 107V, and 107W while being controlled by the microcomputer 130.

Resistors 108U, 108V, 108W are connected to motor windings 107U, 107V, 107W, respectively. As shown in FIG. 1, the resistors 108U, 108V, 108W and the motor windings 107U, 107V, 107W are wired in a contrasting manner, and the middle point 110 of the resistors 108U, 108V, 108W and the motor winding 107U. , 107V, 107W, the midpoint 109 has the same potential.
The control circuit 141 further includes a differential amplifier 103, a DC cutoff filter 102, an integrator 101, and a comparator 104.
The differential amplifier 103 is connected to the middle point 110 of the resistors 108U, 108V, and 108W and the resistor 108U, and outputs a potential difference between both ends of the resistor 108U, that is, a voltage appearing in the resistor 108U. Here, the potentials of the midpoint 109 and the midpoint 110 are the same, and since the current for driving the rotor 112 does not flow through the motor winding 107U as described later, the voltage output from the differential amplifier 103 Is the same as the voltage induced in the motor winding 107U by the rotation of the rotor 112. Hereinafter, the V-phase potential with reference to the midpoint 109 is expressed as Vu-n or the like. The subscript u represents the U-phase terminal and n represents the midpoint 109.

In this embodiment, the drive voltage vectors 3 and 6 are alternately output to rotate the rotor 112. That is, a current is alternately passed through the motor windings 107V and 107W in the V → W direction and the W → V direction, and the motor winding 107U is set to a non-energized phase. When the rotor 112 rotates, an induced electromotive force is generated in the motor winding 107U due to the rotation of the rotor 112. This voltage draws a sine curve as the rotor 112 rotates, and the phase of this sine curve and the position of the magnetic pole of the rotor 112 have a corresponding relationship. Since the voltage generated in the motor winding 107U and the voltage generated in the resistor 108U are the same as described above, the voltage generated in the resistor 108U is detected by the differential amplifier 103, and this signal is processed appropriately. The position of the rotor 112 can be detected.
The voltage generated in the motor winding 107U can be detected by directly connecting the negative terminal of the differential amplifier 103 to the midpoint 109 of the motor 105 instead of the midpoint 110, but the differential is applied to the midpoint 109 due to the motor structure. Since the terminal of the amplifier 103 cannot be connected, a method of detecting the induced electromotive force indirectly by the resistor 108 is employed.

The DC cutoff filter 102 cuts the DC component of the induced electromotive force induced from the differential amplifier 103 to the motor winding 107U. This is because if the output of the differential amplifier 103 includes a direct current component, the integrator 101 integrates it, so that the direct current cutoff filter 102 removes the direct current component in advance. The DC cutoff filter 102 can be realized using a high-pass filter.
The integrator 101 integrates the output of the differential amplifier 103 from which the DC component is removed, and removes electrical noise superimposed on the output of the differential amplifier 103. When the motor operates, various electrical noises are generated. These noises are superimposed on the signal obtained by the differential amplifier 103, and this signal cannot be used as it is. When the signal buried in the noise is integrated by the integrator 101, the noise is averaged, and only the signal buried in the noise can be obtained.
This is because the noise superimposed on the signal is randomly generated at a rate approximately equal to both positive and negative with respect to the signal, so that when integrated, the noise is averaged and canceled.
A signal output from the integrator 101 is referred to as a magnetic flux estimation signal. This is because if the voltage generated in the motor winding is integrated, the interlinkage magnetic flux of the motor winding 107U is obtained.

The input terminal of the comparator 104 is connected to the integrator 101 and the ground, and the output terminal is connected to the microcomputer 130. The comparator 104 outputs a binary signal (a signal in which two types of high and low voltages are associated with each other, Hi indicates a high voltage and Lo indicates a low voltage).
The comparator 104 compares the magnetic flux estimation signal with the ground level, and outputs Hi if the magnetic flux estimation signal is larger than the ground level, and outputs Lo if the magnetic flux estimation signal is smaller than the ground level. The output of the comparator 104 will be referred to as a ROT (rotation pulse signal) signal. Thus, the comparator 104 generates a pulse signal synchronized with the rotor 112.
The microcomputer 130 receives the ROT signal from the comparator 104, switches the transistors 121c, 121d, 121e, and 121f of the motor drive circuit 115 in synchronization with the ROT signal, and supplies a predetermined drive voltage vector to the motor windings 107V and 107W. Output. When the ROT signal is Lo, the transistors 121f and 121c are turned on to output the drive voltage vector 3, and when the ROT signal is Hi, the transistors 121e and 121d are turned on to output the drive voltage vector 6.

The brushless motor control circuit 141 of the present embodiment rotates the rotor 112 by alternately outputting the drive voltage vectors 3 and 6 to the motor windings 107V and 107W among the motor windings 107U, 107V and 107W. Then, the magnetic pole position of the rotor 112 is detected from the voltage induced in the motor winding 107U by the rotation of the rotor 112, and the switching of the drive voltage vectors 3 and 6 is feedback-controlled based on the detection result.
2 shows currents Iu, Iv, Iw flowing through the motor windings 107U, 107V, 107W, an output Vu-n of the differential amplifier 103, a magnetic flux estimation signal φu-n output from the integrator 101, and output from the comparator 104. The relationship between the ROT signal and the drive voltage vectors 3 and 6 is shown.

The operation of the brushless motor control circuit 141 will be described below with reference to the waveform diagram of FIG.
When starting the motor, the drive voltage vector 3, that is, the current flowing in the V → W direction and the drive voltage vector 6, that is, the current flowing in the W → V direction, are alternately repeated at a frequency close to DC, and the magnetic poles of the rotor 112 Is caused to follow the magnetic field generated by the motor winding 107V and the motor winding 107W. When the rotor 112 rotates about once per second, the voltage induced in the motor winding 107U can be detected.
While the drive voltage vector 3 is output, current flows in the direction of V → W, while the drive voltage vector 6 is output, current flows in the direction of W → V, and no current flows in the motor winding 107U. Therefore, the waveforms of the currents Iu, Iv, and Iw are as shown in FIG. Iu is 0.
When the drive voltage vectors 3 and 6 are alternately output and the rotor 112 rotates, the induced electromotive force Vu-n shown in FIG. 2 is generated in the motor winding 107U. As described above, this voltage becomes a sine curve synchronized with the rotation of the rotor 112. This voltage is detected by the differential amplifier 103 connected to both ends of the resistor 108U.
Vu-n output from the differential amplifier 103 is input to the integrator 101 after the DC component is cut by the DC cutoff filter 102.

Next, Vu-n is integrated by the integrator 101 and converted into a magnetic flux estimation signal φu-n. By integrating, the noise superimposed on Vu-n is removed, and the signal buried in the noise can be detected.
Since the magnetic flux estimation signal φu-n is obtained by integrating Vu-n, the phase is delayed by 90 ° as shown in FIG.
The comparator 104 compares the ground level with the magnetic flux estimation signal φu-n and generates an ROT signal. When the magnetic flux estimation signal φu-n is equal to or higher than the ground level, the Lo signal is output, and when it is equal to or lower than the ground level, the Hi signal is output. The ROT signal has a waveform as shown in FIG. The ROT signal is synchronized with the rotation of the rotor 112, and alternately repeats Hi and Lo each time the rotor 112 makes a half rotation.

Next, the microcomputer 130 receives a signal from the comparator, and switches a predetermined transistor of the motor drive circuit 115 based on this signal.
The microcomputer 130 turns on the transistors 121f and 121c of the motor drive circuit 115 while the ROT signal is Lo, and turns on the transistors 121e and 121d while the ROT signal is Hi.
The motor drive circuit 115 outputs the drive voltage vector 3 to the motor 105 while the transistors 121f and 121c are on, and outputs the drive voltage vector 6 to the motor 105 while the transistors 121e and 121d are on.
The current flowing in the motor windings 107V and 107W by the drive voltage vectors 3 and 6 is PWM (pulse width modulation) controlled by the microcomputer 130.
Other drive voltage vectors 1, 2, 4, and 5 are not output.

  As described above, in the present embodiment, the drive voltage vectors 3 and 6 are alternately output to drive the rotor 112, and the magnetic poles of the rotor 112 are derived from the voltage induced in the motor winding 107U that is not used to drive the rotor 112. Detect position. Then, the drive voltage vectors 3 and 6 are switched according to the detected magnetic pole position. Noise is superimposed on the voltage induced in the motor winding 107U. Since the noise is removed by integrating it, the rotation of the rotor 112 can be detected even at a low speed of about 1 rotation per second. Therefore, even when the rotor 112 that cannot lock the PLL circuit is rotated at a low speed, the drive voltage vectors 3 and 6 can be feedback-controlled by the magnetic pole position.

As described above, conventionally, when the motor is started, the drive voltage vector is switched in an open loop, and the drive voltage vector switching time is set to be long so that the rotor can follow the change in the magnetic field. On the other hand, in the present embodiment, the drive voltage vector can be switched quickly as the rotational speed of the rotor increases, so that the motor start time can be shortened.
In addition, since the rotor position can be detected even when the rotor rotates at a low speed, the switching of the drive voltage vector can follow the change in the rotor speed even when the rotor load fluctuates and the rotor speed changes abruptly. Can do. Further, even when a power failure occurs during motor startup, the startup can be resumed without stopping the rotation of the rotor when the power supply is restored.

In the present embodiment, the drive voltage vectors 3 and 6 are used to detect the magnetic pole position by the motor winding 107U. This is because the drive voltage vectors 1 and 4 are used to detect the magnetic pole position by the motor winding 107W. Alternatively, the magnetic pole position may be detected by using the drive voltage vectors 2 and 5 and the motor winding 107V.
When a motor drive is used for the magnetic bearing, the switching noise of the motor driver may propagate to a sensor such as a displacement sensor of the magnetic bearing or a temperature sensor via the bearing body or circuit. In particular, when the magnetic bearing is digitally controlled using a digital signal processor or the like, the displacement signal is detected by sampling with an A / D (analog / digital) converter, so noise is superimposed on the displacement signal of the magnetic bearing. If this is sampled at a moment, a displacement signal including an error is detected, and as a result, noise and vibration may be generated from the magnetic bearing. Therefore, the motor driver used for the magnetic bearing suppresses noise and vibration from the magnetic bearing if the motor is switched off or not switched at the moment of sampling the displacement sensor signal (for example, 2 microseconds). it can.

(Second Embodiment)
A second embodiment of the brushless motor control circuit of the present invention will be described below with reference to FIGS.
In the present embodiment, drive voltage vector 3 and drive voltage vector 5 are alternately output to generate a magnetic field alternately in motor windings 107V and 107W and motor windings 107W and 107U, and rotor 112 is attracted to this magnetic field. Rotate. Then, an ROT signal is generated from the voltage difference between the U-phase terminal and the V-phase terminal, and the drive voltage vectors 3 and 5 are feedback-controlled by this ROT signal.
FIG. 3 is a diagram showing the control circuit 142 according to the present embodiment. Parts having the same functions as those of the control circuit 141 of the first embodiment are denoted by the same numbers.
The difference between the configurations of the control circuit 141 and the control circuit 142 is that the control circuit 142 does not have the star-connected resistors 108U, 108V, and 108W, and that the differential amplifier 103 has a voltage difference between the U-phase terminal and the V-phase terminal. It is a point to detect. The configuration of other parts of the control circuit 141 and the control circuit 142 is the same.

  In the first embodiment, since no motor drive current flows through the motor winding 107U, Vu-n has a clean sine curve. However, in this embodiment, the motor winding 107V in which the motor drive current is flowing. In order to monitor the voltages Vv-n and Vu-n generated at 107U, these voltages include spike-shaped voltages 117 due to the inductances of the motor windings 107U, 107V and 107W, as shown in FIG. 118 etc. appear. In order to eliminate such spike-like voltages 117 and 118, the difference between the two is obtained by the differential amplifier 103. Since voltages Vv-n and Vu-n have spike-shaped voltages 117 and 118 having the same magnitude in the same phase, they can be eliminated by taking the difference.

The configuration of the brushless motor control circuit 142 will be described below.
The motor 105 is composed of motor windings 107U, 107V, 107W connected in a star connection and a rotor 112 having two magnetic poles, N and S.
By alternately outputting the drive voltage vectors 3 and 5, the motor windings 107 </ b> V and W and the motor windings 107 </ b> V and 107 </ b> U alternately generate a magnetic field, and the magnetic poles of the rotor 112 are attracted and rotated. In the first embodiment, since the rotor 112 is rotated by alternately outputting the drive voltage vectors 3 and 6, the rotor 112 may not be started depending on the magnetic pole position of the rotor 112 (drive voltage vector 3, 6). 6 is parallel to each other. Therefore, when the direction of the magnetic field and the direction of the magnetic poles are parallel when the rotor 112 is stopped, no torque is generated in the rotor 112 and the rotor 112 cannot be started. Can be started no matter what position it is at startup.
Also in the first embodiment, it is possible to start the motor by generating a magnetic field other than the drive voltage vectors 3 and 6 for a short time at the start.

The input terminal of the differential amplifier 103 is connected to the motor windings 107U and 107V, and outputs a voltage difference Vu−v between both terminals.
As shown in FIG. 4, spike-like voltages 117, 118, etc. due to the inductances of the motor windings 107U, 107V, 107W appear in Vu-n, Vv-n. Since these have the same magnitude and generated phase, they are eliminated by taking the difference by the differential amplifier 103.
A waveform indicated by a dotted line in Vu-v in FIG. 4 indicates an output from the differential amplifier 103. In this waveform, spike-like voltages 117 and 118 are erased, but still include a DC component 119. This DC component 119 is caused by the resistance values of the motor windings 107U, 107V, and 107W.
The DC cutoff filter 102 cuts the DC component 119, and the waveform shown by the solid line of Vu-v in FIG.

The integrator 101 integrates Vu-v and outputs a magnetic flux estimation signal φu-v. The phase of the magnetic flux estimation signal φu-v is delayed by 90 ° from Vu-v by integration. Further, the noise superimposed on Vu-v is eliminated by integration.
The comparator 104 compares the magnetic flux estimation signal φu-v with the ground level and outputs an ROT signal. Similar to the first embodiment, the ROT signal becomes Hi when the magnetic flux estimation signal is larger than the ground level, and becomes the signal Lo when the magnetic flux estimation signal φ is smaller than the ground level.
The microcomputer 130 turns on and off the transistors 121b, 121c, 121e, and 121f of the motor drive circuit 115 in synchronization with the ROT signal.
The motor driving circuit 115 outputs the driving voltage vector 3 to the motor windings 107V and 107W when the transistors 121c and 121f are on, and outputs the driving voltage vector 5 to the motor windings 107W and 107U when the transistors 121e and 121b are on. Output.

Next, the operation of the control circuit 142 configured as described above will be described.
When the motor is started, a case where a current flows in the driving voltage vector 3, that is, the direction of V → W, and a case where a current flows in the direction of the driving voltage vector 5, ie, the direction of W → U, are alternately repeated at a frequency close to DC (direct current). The rotor 112 is attracted by the magnetic field generated by the motor windings 107V and 107W when the driving voltage vector 3 is output, and by the magnetic field generated by the motor windings 107W and 107U when the driving voltage vector 5 is output. Start spinning. When the rotational speed of the rotor 112 is about 1 rotation per second, the magnetic pole position can be detected.
Currents indicated by Iu, Iv, and Iw in FIG. 4 flow through the motor winding 107U, the motor winding 107V, and the motor winding 107W, respectively. While drive voltage vector 3 is being output, current flows from motor winding 107V to motor winding 107W, and while drive voltage vector 5 is being output, current flows from motor winding 107W to motor winding 107U.

A waveform Vu-n in FIG. 4 represents the voltage of the U-phase terminal with the midpoint 109 as a reference. As described above, the spike-like voltage 117 appearing at various points in the waveform is due to the influence of a voltage drop due to the inductance of the motor winding 7.
A waveform Vv-n in FIG. 4 represents the voltage at the V-phase terminal with the midpoint 109 as a reference. In the waveform, spike-like voltage 118 appears in some places for the same reason as Vu-n.
The spike-like voltages 117 and 118 appearing in these two waveforms have the same phase and the same magnitude.
These Vu-n and Vv-n are input to the differential amplifier 103, and the difference between them is output.
A waveform indicated by a dotted line in the waveform Vu-v in FIG. 4 represents an output signal of the differential amplifier 103. As described above, the spike-like voltages 117 and 118 appearing at Vv−n and Vu−n are equal in magnitude to the place where they are generated.

The output signal of the differential amplifier 103 is superimposed with a DC component 119, which is cut by the DC cutoff filter 102. This is to prevent the DC component from being integrated by the integrator 101. From the DC cutoff filter 102, a Vu-v waveform shown by a solid line in FIG.
Vu-v from which the DC component has been cut by the DC cutoff filter 102 is integrated by the integrator 101 and converted to the magnetic flux estimation signal φu-v shown in FIG. The noise superimposed on Vu-v by the integration is removed, and only the target signal is obtained. The magnetic flux estimation signal φu-v changes in synchronization with the rotor 112.
The magnetic flux estimation signal φu-v output from the integrator 101 is compared with the ground level by the comparator 104, and the ROT signal (rotation pulse signal) shown in FIG. 4 is output. The ROT signal becomes Lo when the magnetic flux estimation signal φu-v is larger than the ground level, and becomes Hi when it is smaller than the ground level.

The microcomputer 130 receives the ROT signal from the comparator, turns on the transistors 121f and 121c of the motor drive circuit 115 while the ROT signal is Lo, and turns on the transistors 121e and 121b while the ROT signal is Hi.
As shown in FIG. 4, when the ROT signal is Lo, the drive voltage vector 5 is output to the motor windings 107V and 107W, and when the ROT signal is Hi, the drive voltage vector 3 is output to the motor windings 107W and 107U. The
The current flowing in the motor windings by these drive voltage vectors 3 and 5 is PWM controlled by the microcomputer 130. The drive voltage vectors 1, 2, 4, and 6 are not output.

In the present embodiment, it is possible to start without fail regardless of the magnetic pole stop position of the rotor 112. This is because the magnetic field generated by the drive voltage vectors 3 and 5 is not parallel, and torque is generated in the rotor regardless of the position where the magnetic pole is stopped.
In addition, the control circuit 142 according to the present embodiment does not require a resistor for detecting the voltage of the motor winding, and the PLL circuit cannot be locked because the rotor 112 rotates at a low speed with a simple circuit configuration. However, the position of the rotor 112 can be detected and the switching of the drive voltage vector can be feedback controlled.
The motor starting time can be shortened, and the motor 105 can be controlled without stepping out even when the load on the rotor 112 fluctuates or the power supply is restored after a power failure. The same.
In the present embodiment, the motor is driven using the drive voltage vectors 3 and 5, but the method of selecting the drive voltage vector is not limited to this. For example, the drive voltage vectors 5 and 1 or the drive voltage vectors 1 and 3 are It may be used. That is, the magnetic flux estimation signal can be obtained by selecting the drive voltage vector so that the spike-like voltage appearing due to the inductance of the motor winding 7 is canceled by the differential amplifier 103.

(Third embodiment)
In this embodiment, after the number of rotations of the rotor of the motor activated by the means of the second embodiment reaches the number of rotations that can lock the PLL circuit (about 20 rotations per second), steady operation using the PLL circuit is performed. A case of switching to will be described.
During the steady operation of the motor, while detecting the position of the magnetic pole, the PLL circuit generates 6 pulses every time the rotor makes one rotation, and the drive voltage vector is changed from 1 → 2 → 3 → 4 → 5 → in synchronization with this pulse. The rotor is rotated by sequentially switching in the order of 6 → 1 →.
On the other hand, the rotation of the rotor 112 of the second embodiment is performed by alternately switching the drive voltage vectors 3 and 5.
Therefore, when switching from the operation by the means of the second embodiment to the normal operation using the PLL circuit, it is necessary to appropriately obtain the drive voltage vector to be output first and the timing to output this drive voltage vector. If an inappropriate drive voltage vector is output when switching to normal operation, or if the output timing of the drive voltage vector is incorrect, the motor will step out or generate an abnormal sound.

FIG. 5 shows a control circuit 143 of the brushless motor according to the present embodiment. The control circuit 143 further includes a rotation speed sensor 125, a rotation speed detection circuit 126, a PLL circuit 127, and a storage unit 128 in addition to the brushless motor control circuit 142 according to the second embodiment. Parts having the same functions as those of the brushless motor control circuit 142 according to the second embodiment are denoted by the same reference numerals.
The rotation speed sensor 125 is attached to the outside of the motor 105, detects a target attached to the rotor shaft, and generates a signal synchronized with the rotation of the rotor 112. For example, the rotation speed sensor 125 is formed by a Hall element or the like, and the target is configured by a magnet or the like. The rotational speed detector may be installed inside the motor 105 to directly detect the magnetic poles of the rotor 112. From the rotation speed sensor 125, it is possible to obtain a clean waveform on which no electrical noise is superimposed.

The rotation speed sensor 125 is connected to the rotation speed detection circuit 126. The rotation speed detection circuit 126 calculates the rotation speed of the rotor 112 from the output signal of the rotation speed sensor 125 and outputs a pulsed ROTA signal (second rotation pulse signal). The ROTA signal is a binary signal (a signal in which two types of high and low voltages are associated with each other. Hi is a signal having a high voltage and Lo is a signal having a low voltage).
The ROTA signal becomes Hi during the half rotation of the rotor 112, and becomes Lo during the subsequent half rotation.
The microcomputer 130 and the PLL circuit 127 are connected to the rotational speed detection circuit 126 and receive the ROTA signal from the rotational speed detection circuit 126.
The PLL circuit 127 generates a pulse-like synchronization signal 6 × fROTA signal synchronized with a frequency that is six times the frequency of the ROTA signal. This synchronization signal is used when switching the six drive voltage vectors.

From the ROT signal and the ROTA signal, the microcomputer 130 calculates the timing for outputting the drive voltage vector to be output first when switching to the normal operation and the synchronization signal generated by the PLL circuit 127.
A storage unit 128 is connected to the microcomputer 130. This is because when the rotational speed of the rotor 112 is increased, the output signal of the rotational speed sensor 125 may be delayed from the timing at which the target reaches the detection position, and a value for correcting this is stored in the storage unit 128 in advance. The microcomputer 130 corrects the rise time and fall time of the ROTA signal using this correction value in accordance with the rotational speed of the rotor 112.
The microcomputer 130 performs a predetermined calculation from these values, and the gate drive pulse to the transistors 121a, 121b, 121c, 121d, 121e, and 121f of the motor drive circuit 115 is obtained from the start operation of the second embodiment. Switch to normal operation.

The operation of the brushless motor control circuit 143 configured as described above will be described.
The motor 105 is driven by the method described in the second embodiment until the rotation speed of the rotor 112 reaches a rotation speed (for example, 20 rotations per second) that can be locked by the PLL circuit after the motor 105 is started. That is, the drive voltage vectors 3 and 5 are alternately output, the ROT signal is generated by integrating the voltage difference between the V-phase terminal and the U-phase terminal, and the drive voltage vectors 3 and 5 are synchronized with the ROT signal. Switch.
When the number of rotations of the rotor 112 reaches the minimum number of rotations that can lock the PLL circuit, the microcomputer 130 calculates the drive voltage vector to be output first and the timing to output it at the time of switching the normal operation, and the motor 105 is normally operated. Switch to driving.
The drive voltage vector that is output first at the time of operation switching is calculated as follows.
The microcomputer 130 detects and stores the period T of the ROT signal. Further, the time difference Ta between the ROT signal and the ROTA signal is detected and stored.
In FIG. 6, ROT represents the ROT signal, and ROTA represents the ROTA signal. Φu−v is a magnetic flux estimation signal φu−v output from the integrator 101.
Next, the phase difference Y between the ROT signal and the ROTA signal is calculated by the following equation.

(Equation 1)
Y = Ta / T (1)

FIG. 7 shows the value of Y and the drive voltage vector number output first when switching to normal operation.
For example, when 3/12 ≦ Y <5/12, the driving voltage vector 1 is started.
When the drive voltage vector is output according to the table in FIG. 7, the position of the magnetic pole of the rotor 112 and the magnetic field generated by the drive voltage vector are in an appropriate positional relationship.
The drive voltage vector number and the waveform of the magnetic flux estimation signal φu-v in the column “Voltage vector to be driven in the three-phase full-wave mode” in FIG.
Since φu-v and the ROT signal are synchronized, the correspondence relationship between the ROT signal and the drive voltage vector can also be seen from this waveform diagram. For example, the rise of the ROT signal is the middle of the period during which the drive voltage vector 4 is output.
One cycle of the ROT signal may be divided into six and drive voltage vectors may be output in the order of 1 → 2 → 3 →... In synchronization with the ROT signal.
Since the waveform of the ROTA signal is obtained by translating the waveform of the ROT signal, the ROTA signal and the number of the drive voltage vector can be made to correspond to each other as shown in FIG.

Next, the microcomputer 130 calculates how much the 6 × fROTA signal, which is a 6-fold synchronization signal of the ROTA signal, should be output after being delayed from the ROT signal.
Since the number of the drive voltage vector to be output corresponding to the ROTA signal is obtained by Equation 1, the output timing of the 6 × fROTA signal is finely adjusted, and the rise and fall timings of the 6 × fROTA signal are set to the ROT signal rise time. Align with the falling timing.
The “phase delay amount D” column in FIG. 7 shows the phase delay amount corresponding to each Y value. The microcomputer 130 calculates the phase delay time Td from the period T of the ROTA signal and the phase delay amount D by the following equation.

(Equation 2)
Td = T × D (2)

The microcomputer 130 delays the 6 × fROTA signal generated by the PLL circuit 127 by Td and outputs a gate drive pulse corresponding to the vector number to be driven to the motor drive circuit 115. Each time the 6 × fROTA signal rises, the gate drive pulse updates the drive voltage vector in order of 1 → 2 → 3 → 4 →.
Further, when the rotational speed of the rotor 112 increases, the detection signal of the rotational speed sensor 125 may be delayed from the time when the target reaches the detection position of the rotational speed sensor 125. In this case, the ROTA signal is output with a delay from the value that should be originally. If the relationship between the amount of delay and the rotation speed of the rotor 112 is known in advance, these are stored in the storage unit 128 as correction values, and the ROTA signal is corrected inside the microcomputer 130.

By the above means, the motor 105 started by the means of the second embodiment can be promptly shifted to the steady state by the 6 × fROTA signal of the PLL circuit 127.
Normally, control of a sensorless brushless motor requires measuring the impedance of the motor wiring or monitoring and correcting the motor current. However, the brushless motor control circuit 143 of the present embodiment does not need these. Therefore, the circuit becomes simple and the cost is reduced.
Further, when the motor 105 started by the brushless motor control circuit 141 of the first embodiment is switched to the steady operation, it can be performed in the same manner as in the present embodiment.

(Fourth embodiment)
In the present embodiment, the magnetic pole position of the rotor is detected from a change in magnetic flux generated in the motor winding.
First, a theoretical formula that is fundamental for carrying out this embodiment is derived.
FIG. 10 is a diagram showing the resistance value and inductance of the cable connecting the motor and the motor winding. The U-phase, V-phase, and W-phase motor windings are star-connected.

Viu-n is a voltage generated in the U-phase motor winding by the rotation of the rotor. The U-phase motor winding has an inductance Lu and a resistance value Ru. The cable has a wiring resistance value Rc. Since the V phase and the W phase are structurally symmetric with respect to the U phase, the above subscripts u are set to v and w, respectively, so as to correspond to the V phase and the W phase, respectively. In addition, currents flowing in the U phase, the V phase, and the W phase are Iu, Iv, and Iw, respectively.
Here, Rp, Lp, and Viu-v are set as follows.

(Equation 3)
Rc + Ru = Rc + Rw = Rc + Rv = Rp
Lu = Lw = Lv = Lp
(Viu-n)-(Viv-n) = Viu-v

  The following equation is derived from the above equation.

(Equation 4)
Vu−v = Viu−v + Rp (Iu−Iv) + Lp × d (Iu−Iv) / dt

  Here, Vu-v is obtained by subtracting the voltage of the V-phase motor winding from the voltage of the U-phase motor winding. When this is integrated, the following equation (3) is obtained.

(Equation 5)
∫Vu−vdt = ∫ (Viu−v + Rp × (Iu−Iv)) dt + Lp × (Iu−Iv) (3)

  Here, the magnetic flux estimation signal φu−v indicating the magnetic flux in the U-phase motor winding and the V-phase motor winding is expressed by the following equation (4).

(Equation 6)
φu−v = ∫Viu−vdt (4)

  Here, various quantities represented by the following formulas (5), (6), and (7) are considered.

∫Vu-vdt (5)
∫Rp × (Iu−Iv) dt (6)
Lp × (Iu−Iv) (7)

It can be seen from equation (3) that the magnetic flux φu−v can be obtained by subtracting equations (6) and (7) from equation (5).
In the present embodiment, these relational expressions are realized by using an electric circuit, the magnetic flux estimation φu−v is obtained, and the motor winding current is switched based on the magnetic flux estimation signal φu−v.

FIG. 11 is a block diagram showing the configuration of the control circuit 41 of the sensorless brushless motor according to the fourth embodiment of the present invention.
The control circuit 41 includes a motor 5, a motor drive circuit 17, and a drive control circuit 19.
The motor 5 includes a motor winding 7U, 7V, and 7W that are star-connected, and a rotor 6 that has a pair of magnetic poles, an N pole and an S pole. Although the motor windings 7U, 7V, and 7W and the rotor 6 are shown separately for convenience of illustration, the motor windings 7U, 7V, and 7W are actually arranged around the rotor 6. When driving the motor, for example, current is passed through the motor windings 7U and 7V in the direction of U → V, and current is passed through the two motor windings. Let this rotate. The rotation of the rotor 6 is maintained by sequentially switching the motor winding through which the current flows and the direction of the current based on the position of the magnetic poles of the rotor 6.
The motor drive circuit 17 includes a DC power supply 18 and six transistors 21a, 21b, 21c, 21d, 21e, and 21f that form a three-phase bridge. The bases of these transistors are respectively connected to a microcomputer 30 in the drive control circuit 19, and are turned on / off by a gate signal from the microcomputer 30 to supply a predetermined current to the motor windings 7U, 7V, 7W.

In addition, the control circuit 41 includes circuit elements that calculate the magnetic flux estimation signal φu−v by performing calculations from the above equations (4) to (7). These circuit elements are composed of differential amplifiers 8 and 9, multipliers 10 and 12, adders 11 and 13, integrator 1, and DC cutoff filter 2.
The differential amplifier 8 is connected to the motor windings 7U and 7V, and outputs a value Vu−v obtained by subtracting the voltage of the motor winding 7V from the voltage of the motor winding 7U.
The differential amplifier 9 detects currents Iu and Iv flowing through the motor windings 7U and 7V and outputs a difference Iu-Iv between them.
The multiplier 10 is connected to the differential amplifier 9 and the Rp signal setting circuit 14 in the drive control circuit 19. The Rp signal setting circuit 14 has a resistance value of any one of the motor windings 7U, 7V, and 7W (the resistance value is the same for all three) and cables 3U, 3V, and 3W that connect the motor winding and the motor drive circuit 17. The combined resistance value of any one of the above (the resistance value is the same for all three). The multiplier 10 receives Iu−Iv from the differential amplifier 9 and Rp from the Rp signal setting circuit 14 and outputs Rp × (Iu−Iv) which is the product of both.
The adder 11 is connected to the differential amplifier 8 and the multiplier 10. The adder 11 receives Vu−v from the differential amplifier 8, Rp × (Iu−Iv) from the multiplier 10, and from Vu−v. A difference obtained by subtracting Rp × (Iu−Iv), that is, (Vu−v) −Rp × (Iu−Iv) is output. This is the integrand of equation (6).

The output of the adder 11 is input to the integrator 1 via the DC cutoff filter 2. This is to prevent the DC component included in the output of the adder 11 from being integrated by the integrator 1.
The integrator 1 integrates the output of the adder 11 from which the DC component has been removed, and outputs ∫ ((Vu−v) −Rp × (Iu−Iv)) dt. The output of the integrator 1 corresponds to the case where the equation (6) is subtracted from the equation (5). Further, by integrating, the electric noise superimposed on the input signal of the integrator 1, that is, (Vu−v) −Rp × (Iu−Iv), is removed. This is because noise is randomly generated positively and negatively with respect to the signal value, and is canceled by integrating and adding them.
The multiplier 12 is connected to the differential amplifier 9 and the Lp signal setting circuit 15. The multiplier 12 receives Iu−Iv from the differential amplifier 9 and Lp from the Lp signal setting circuit 15, and the product of both, ie, Lp × (Iu−Iv ) Is output. This value corresponds to equation (7).
The adder 13 outputs the output of the integrator 1, that is, the value obtained by subtracting the output of the multiplier 12, that is, Lp × (Iu−Iv) from ∫ ((Vu−v) −Rp × (Iu−Iv)) dt. . The output of the adder 13 corresponds to a value obtained by subtracting Expression (6) and Expression (7) from Expression (5), and is equal to the magnetic flux estimation signal φu−v. The waveform of the magnetic flux estimation signal φu-v becomes a sine curve synchronized with the rotation of the rotor 6.

The drive control circuit 19 includes a comparator 4, a PLL (Phase Lock Loop) circuit 16, an Lp signal setting circuit 15, an Rp signal setting circuit 14, and a microcomputer 30. The comparator 4 has two input terminals, one connected to the magnetic flux estimation signal φu-v and the other connected to the ground. The comparator 4 outputs a binary signal (a signal in which two kinds of high and low voltages are associated with each other, Hi is a signal having a high internal voltage, and Lo is a low voltage).
The comparator 4 compares the magnetic flux estimation signal φu-v with the ground level, and outputs Lo when the magnetic flux estimation signal φu-v is smaller than the ground level, and outputs Hi when the magnetic flux estimation signal φu-v is larger than the ground level. To do.
Since φu−v becomes a sine wave synchronized with the rotor 6, the comparator 4 outputs Hi during the half rotation of the rotor and Lo during the subsequent half rotation. This signal is called an ROT signal (rotation pulse signal). The output terminal of the comparator 4 is connected to the microcomputer 30 and the PLL circuit 16.

The PLL circuit 16 receives the ROT signal, generates a 12 × fROT signal that is a multiplied synchronization signal synchronized with a frequency 12 times the rotational speed of the rotor 6, and outputs the signal to the microcomputer 30.
The Rp signal setting circuit outputs the resistance value Rp stored in the microcomputer 30 to the multiplier 10.
The Lp signal setting circuit outputs the inductance value Lp stored in the microcomputer 30 to the multiplier 12.
The resistance value of the motor windings 7U, 7V, 7W and the combined resistance value Rp of the resistance values of the cables 3U, 3V, 3W connecting this to the motor drive circuit 17 and the inductance Lp of the motor winding are measured in advance using a measuring instrument. It is measured and stored in the microcomputer 30.
The microcomputer 30 supplies a predetermined gate signal to the transistors 21a, 21b, 21c, 21d, 21e, and 21f based on the 12 × fROT signal of the PLL circuit 16, and sequentially switches the currents of the motor windings 7U, 7V, and 7W. It has become.

The operation of the control circuit 41 of the sensorless brushless motor configured as described above will be described with reference to the waveform diagram of FIG.
Since the frequency at which the PLL circuit 16 can be locked (operated) is about 20 [Hz], the motor windings 7U and 7V are opened in an open loop until the rotational speed of the rotor 6 reaches about 20 revolutions per second after the motor 5 is started. , Switch the current of 7W.
When the rotational speed of the rotor 6 reaches about 20 revolutions per second, an ROT signal is generated from the magnetic flux estimation signal φu-v, and thereby, the current switching of the motor windings 7U, 7V, 7W can be feedback controlled. .
The currents of the motor windings 7U, 7V, and 7W have waveforms of Iu, Iv, and Iw in FIG. The microcomputer 30 performs PWM (Pulse Width Modulation) control on the voltage applied to the motor windings 7U, 7V, and 7W so that the current becomes a rectangular wave.
The differential amplifier 8 takes the voltage difference between the motor windings 7U and 7V and outputs the waveform indicated by Vu-v in FIG. The spike-like voltage 20 that appears when the currents of the motor windings 7U, 7V, and 7W are switched is due to the inductance Lp of the motor windings 7U, 7V, and 7W, and the step 21 between the adjacent waveforms is the motor winding 7U. , 7V, 7W resistance values and the resistance values of the cables 3U, 3V, 3W connecting these to the motor drive circuit 17 are voltage drops due to the combined resistance value Rp.

The differential amplifier 9 outputs the difference between Iu and Iv, and the waveform thereof is the waveform indicated by Iu-Iv in FIG.
Multiplier 10 multiplies Iu-Iv and Rp.
The adder 11 outputs (Vu−v) −Rp × (Iu−Iv). This signal is input to the integrator 1 after the superimposed DC component is removed by the DC cutoff filter 2.
The integrator 1 outputs ∫ ((Vu−v) −Rp × (Iu−Iv)) dt, which has a waveform indicated by X in FIG. The step observed in the waveform X is caused by the inductance of the motor windings 7U, 7V, and 7W.
The output of the integrator 1 is subtracted by Lp × (Iu−Iv) by the adder 13. The adder 13 outputs a magnetic flux estimation signal φu-v indicated by the waveform of φu-v in FIG.
As can be seen from this waveform, the magnetic flux estimation signal φu-v becomes a sine wave having one cycle when the rotor 6 makes one rotation.
The comparator 4 compares the magnetic flux estimation signal φu-v with the ground level and outputs an ROT signal. As described above, the ROT signal becomes Hi when the magnetic flux estimation signal φu-v is smaller than the ground level, and becomes the signal Lo when larger than the ground level. The ROT signal becomes Hi during half rotation of the rotor 6 as shown by the ROT waveform in FIG. 12, and becomes Lo during the subsequent half rotation.

The PLL circuit 16 receives the ROT signal from the comparator 4 and generates a multiplied synchronization signal 12 × fROT signal having a frequency 12 times that of the ROT signal.
There are six ways to switch the current of motor windings 7U, 7V, and 7W, for example, W → V direction, U → V direction, etc., so basically a multiple synchronization signal with 6 times the frequency of ROT signal If the 6 × fROT signal is generated, the six currents can be switched in synchronization with this signal. Here, the reason why the 12 × f ROT signal having a frequency 12 times that of the ROT signal is generated is to correct the phase shift of φu−v due to the integration.
More detailed description is as follows.
The microcomputer 30 switches the current of the motor windings 7U, 7V, 7W every time the rotor 6 rotates 60 °. The vertical line in FIG. 12 is the timing for switching the current, and the interval is 60 ° when the rotor 6 is rotated. By the way, the phase of φu−v advances by 90 ° from the original signal by integration. For this reason, the phase of the ROT signal generated from φu-v is also advanced by 90 °, and there is a deviation of 30 ° from the timing at which the ROT signal rises and falls and the current is switched. Therefore, when a 12 × fROT signal having a period 12 times that of the ROT signal is generated, the timing for switching the current can coincide with the rising edge of the 12 × fROT signal.
In this way, the timing of switching the currents of the motor windings 7U, 7V, 7W and the rising edge of the 12 × fROT signal coincide with each other, and each time the 12 × fROT signal rises twice, If switched, the motor 5 can be operated.

  The microcomputer 30 stores a program for supplying a predetermined gate signal to the bases of the transistors 21a, 21b, 21c, 21d, 21e, and 21f by using the ROT signal and 12 × fROT, and 21a, 21b, and 21c in FIG. , 21d, 21e, and 21f, the transistors corresponding to the respective numbers are switched on. For example, when the transistors 21d and 21e are on, current flows in the W → V direction through the motor windings 7V and 7W. As described above, the current flowing through the motor windings 7U, 7V, and 7W is PWM controlled to a predetermined value by the microcomputer 30.

In this embodiment, since Rp and Lp are measured in advance, an amplifier having a gain Rp can be used instead of the multiplier 10, and an amplifier having a gain Lp can be used instead of the multiplier 12.
In the present embodiment, the current and voltage of the motor windings 7U and 7V are monitored. However, the present invention is not limited to this, and the voltage and current of any motor winding may be monitored.
Since the control circuit 41 of the sensorless brushless motor according to the present embodiment can constantly monitor the magnetic pole position of the rotor 6, the motor 5 can be appropriately controlled without stepping out even when the rotational speed of the rotor 6 changes greatly due to load fluctuations. can do. Further, since the signal for detecting the magnetic pole position of the rotor 6 is integrated, the magnetic pole position can be accurately detected without being affected by the electrical noise superimposed on this signal.

(Fifth embodiment)
In the fourth embodiment, the resistance value of the motor windings 7U, 7V, 7W and the combined resistance value Rp of the resistance values of the cables 3U, 3V, 3W connecting these and the motor drive circuit 17 are measured in advance by a measuring instrument, These are stored in the microcomputer 30. In this embodiment, a case where Rp is automatically measured and stored in the microcomputer 30 will be described.
FIG. 13 is a block diagram showing the configuration of the control circuit 42 of the sensorless brushless motor according to the fifth embodiment of the present invention.
The control circuit 42 of the present embodiment is obtained by newly adding low-pass filters 22 and 23 to the control circuit 41 of the fourth embodiment. Since the other configuration is the same as that of the control circuit 41, these components are given the same numbers as those of the control circuit 41.

The low-pass filter 22 is connected to the differential amplifier 9 and a microcomputer 30 (not shown) in the drive control circuit 19.
The low-pass filter 22 receives Iu-Iv from the differential amplifier 9, removes high-frequency noise superimposed on it, and outputs it to the microcomputer 30.
The low-pass filter 23 is connected to the differential amplifier 8 and a microcomputer 30 (not shown) in the drive control circuit 19. The low-pass filter 23 receives the line voltage Vu-v of the cables 3U and 3V from the differential amplifier 8, removes the high frequency noise superimposed thereon, and outputs it to the microcomputer 30.
The low-pass filters 22 and 23 are used in this way for the signal Iu-Iv output from the differential amplifier 9 and the signal Vu-v output from the differential amplifier 8 in the motor windings 7U, 7V, and 7W. This is because high-frequency noise is superimposed for current switching, PWM control, and the like, which is removed to improve measurement accuracy.

The operation of the control circuit 42 of the sensorless brushless motor configured as described above will be described.
Only the procedure for automatically measuring Rp will be described here. Other operations of the control circuit 42 are the same as those of the control circuit 41 of the fourth embodiment.
When the motor 5 is started, a direct current is passed through the motor windings 7U and 7V in the U → V direction, and the line voltage of the cables 3U and 3V is measured by the differential amplifier 8.
After the noise is removed from the voltage by the low-pass filter 23, the value is stored in the microcomputer 30.
Next, the current value flowing through the cables 3U and 3V is measured by the differential amplifier 9, high frequency noise superimposed on the value is removed by the low-pass filter 22, and the value is stored in the microcomputer 30.
The microcomputer 30 calculates Rp by the following equation.

(Equation 7)
Rp = (Vu−v) / (Iu−Iv) (8)

The microcomputer 30 stores Rp obtained by the equation (8) and outputs it to the Rp signal setting circuit 14 when the motor 5 is operated. This Rp is multiplied by Iu-Iv by the multiplier 10 and used to obtain the magnetic flux estimation signal φu-v.
Note that the measurement accuracy can be improved by keeping the transistor 21d switched on during Rp measurement and performing PWM control only on the transistor 21a.
Here, Iu−Iv is obtained by the differential amplifier 9, but this may be configured such that each of Iu and Iv is detected and the difference is obtained by the microcomputer 30.
Moreover, if the Rp value obtained as a result of the calculation exceeds the normal value range (for example, 0.5 [Ω] to 10 [Ω]), an alarm is output and a safety device is installed so that the motor 5 is not driven. Safety can be increased.

In the present embodiment, a combined resistance value of the resistance values of the motor windings 7U, 7V, and 7W and the resistance values of the cables 3U, 3V, and 3W is automatically measured when the motor 5 is started, and the value is set as Rp. Therefore, unlike the conventional sensorless brushless motor control circuit, it is not necessary to measure Rp in advance using a measuring instrument.
Therefore, conventionally, when the cables 3U, 3V, 3W are extended at the site where the sensorless brushless motor is used, or when the motor 5 is replaced with another motor, it is necessary to remeasure Rp again using a measuring instrument. The control circuit 42 according to the present embodiment does not require such work.

(Sixth embodiment)
In the fourth embodiment, the inductances Lp of the motor windings 7U, 7V, and 7W are measured in advance by a measuring instrument and stored in the microcomputer 30, but in this embodiment, Lp is automatically measured. A case where data is stored in the microcomputer 30 will be described.
FIG. 14 is a block diagram showing the configuration of the control circuit 43 of the sensorless brushless motor according to the sixth embodiment of the present invention.
In the control circuit 43 of the present embodiment, the low-pass filters 22 and 23 of the control circuit 42 of the fifth embodiment are particularly low-pass filters 24 and 25 for removing high frequencies. Other configurations are the same as those of the control circuit 42, and these same components are denoted by the same reference numerals as those of the control circuit 42.
In this embodiment, in order to obtain the inductance of the motor windings 7U, 7V, 7W, a high-frequency AC voltage of about 1 [kHz] is used, for example. Further, since this AC voltage is PWM controlled by a microcomputer 30 (not shown) in the drive control circuit 19, a noise of a PWM frequency, for example, 50 [kHz] is superimposed. Therefore, for example, if a low-pass filter having a cutoff frequency of 5 [kHz] is used, noise due to PWM control can be removed from the target signal, and measurement accuracy can be improved.

The operation of the control circuit 43 of the sensorless brushless motor configured as described above will be described.
While the rotor 6 is stopped, a high-frequency AC voltage Vu-v (for example, ft = 1 [kHz]) at which the rotor 6 cannot rotate in the U → V direction is applied to the motor windings 7U and 7V. The rotor 6 cannot follow the reversing magnetic field created by the motor windings 7U and 7V and does not rotate.
Next, the differential amplifier 9 detects the alternating current Iu-Iv at this time. This value is stored in a microcomputer 30 (not shown) in the drive control circuit after high frequency noise is removed by the low pass filter 25.
On the other hand, the differential amplifier 8 detects the line voltage of the cables 3U and 3V. This value is stored in the microcomputer 30 after high frequency noise is removed by the low pass filter 24.
The microcomputer 30 calculates Lp from the stored AC voltage Vu-v and AC current Iu-Iv by the following equation (9).

(Equation 8)
Lp = (Vu−v) / (2 × π × ft × (Iu−Iv)) (9)

The microcomputer 30 stores the value obtained by the equation (9) as Lp. Then, this is output to the Lp signal setting circuit 15 when the motor 5 is operated. This Lp is multiplied by Iu−Iv by the multiplier 12 and used to obtain the magnetic flux estimation signal φu−v. Other operations of the control circuit 43 are the same as those in the fourth embodiment.
If the Lp value obtained as a result of the calculation exceeds a normal value range (for example, 0 [mH] to 1 [mH]), an alarm is output, and it is safe to install a safety device that does not drive the motor 5. Can be increased.

In the present embodiment, the inductance of the motor windings 7U, 7V, and 7W is automatically measured when the motor 5 is stopped, and the value is output to the Lp signal setting circuit 15 as Lp. It is not necessary to measure Lp using a measuring instrument in advance like a circuit.
Therefore, in the past, when the motor 5 was replaced with another motor at the site where the sensorless brushless motor was used, it was necessary to remeasure Lp again using a measuring instrument. In the control circuit 43 of the present embodiment, however, Such work is unnecessary.
Also, the configuration of the control circuit 43 can measure Rp similarly to the fifth embodiment, and both Rp and Lp can be automatically measured.
Further, since the Lp of a sensorless brushless motor used for a magnetic bearing type turbo molecular pump that normally rotates at a high speed of 300 rpm or more is small, detection of Lp is omitted, and this is set to a predetermined value (for example, several hundred [μH ]) And the motor 5 can be operated even if this is used.

(Seventh embodiment)
In the present embodiment, a case will be described in which the inductance Lp of the motor windings 7U, 7V, and 7W is automatically measured by means different from the sixth embodiment.
FIG. 15 is a block diagram showing the configuration of the control circuit 44 of the sensorless brushless motor according to the seventh embodiment of the present invention. The control circuit 44 is obtained by adding a sampling circuit 26 to the control circuit 41 of the fourth embodiment. The sampling circuit 26 is connected to the integrator 1 and a microcomputer 30 (not shown) in the drive control circuit. When the sampling circuit 26 receives a sample signal output command from the microcomputer 30, the sampling circuit 26 samples the value of the output X of the integrator and transmits this sample signal to the microcomputer 30.
Since other configurations are the same as those of the control circuit 41, the same components as those of the control circuit 41 are given the same numbers.

Next, a theoretical formula that is fundamental when the present embodiment is implemented will be described.
The output X of the integrator 1 sampled by the sampling circuit 26 is expressed by the following equation (10).

(Equation 9)
X = Lp × (Iu−Iv) + ∫Viu−vdt (10)

  Here, when the currents of the motor windings 7U, 7V, and 7W are switched, Lp × (Iu−Iv) of the first term of Expression (10) changes greatly, but the change of ∫Viu−vdt of the second term is small. Further, since the drive control circuit 19 performs PWM control so that the currents Iu, Iv, and Iw supplied to the motor 5 have a constant value Ip, the points on the waveform diagram of X in FIG. X (point 32) −X (point 31), which is the difference between the two points 31 and 32, is expressed by the following equation (11).

(Equation 10)
X (point 32) −X (point 31) = Lp × Ip (11)

Here, since the control circuit 44 only adds the sampling circuit 26 to the control circuit 41, the waveform diagram of the control circuit 44 is the same as that of the control circuit 41.
Lp is represented by Formula (12) from Formula (11).

(Equation 11)
Lp = (X (point 32) −X (point 31)) / Ip (12)

  Further, the measurement accuracy can be improved by calculating the difference between the two points represented by the points 33 and 34 on the X waveform in FIG. In the present embodiment, the average value expressed by the equation (13) is calculated by the microcomputer 30 to obtain Lp.

(Equation 12)
Lp = (X (point 32) −X (point 31) + X (point 33) −X (point 34)) / (2 × Ip) (13)

The operation of the control circuit 44 of the sensorless brushless motor configured as described above will be described with reference to the waveform diagram of FIG. In addition, the part which overlaps with 4th Embodiment is abbreviate | omitted.
The microcomputer 30 outputs a sample signal output command to the sampling circuit 26 at the timing of the point 31, that is, immediately before switching the current in the U → V direction to the motor windings 7 U and 7 V. The sampling circuit 26 outputs the output of the integrator 1. Sampling X (point 31). The microcomputer 30 receives X (point 30) from the sampling circuit 26 and stores it.
Similarly, the microcomputer 30 outputs the output X (point 32) of the integrator 1 via the sampling circuit 26 at the timing of point 32, that is, immediately after switching the current to the motor windings 7U and 7V in the U → V direction. Is received and memorized.
However, immediately after switching the currents of the motor windings 7U, 7V, and 7W, noise appears in the output X of the integrator 1. Therefore, a short time (for example, 50 μsec) until the noise disappears is reduced by the internal timer of the microcomputer 30. After counting, X (point 32) is sampled.
Similarly, X (point 33) and X (point 34) are sampled and those values are stored in the microcomputer 30.
The set value of Ip is stored in the microcomputer 30.

Next, the microcomputer 30 substitutes the stored output X (point 31), X (point 32), X (point 33), and X (point 34) of the integrator 1 into the equation (13) to calculate Lp. The microcomputer 30 stores this Lp. Then, this is output to the Lp signal setting circuit 15 when the motor 5 is operated. This Lp is multiplied by Iu−Iv by the multiplier 12 and used to obtain the magnetic flux estimation signal φu−v. Other operations of the control circuit 43 are the same as those in the fourth embodiment.
In the control circuit 44, since Iu and Iv are monitored by the differential amplifier 9, these may be used as Ip.
If the Lp value obtained as a result of the calculation exceeds a normal value range (for example, 0 [mH] to 1 [mH]), an alarm is output, and it is safe to install a safety device that does not drive the motor 5. Can be increased.

  In the present embodiment, it is not necessary to measure the inductance of the motor windings 7U, 7V, 7W in advance, and it is necessary to remeasure the inductance of the motor winding even when the motor 5 is replaced with another motor. This is the same as the sixth embodiment.

(Eighth embodiment)
In this embodiment, the resistance values Rp of the motor windings 7U, 7V, 7W, the resistance values of the cables 3U, 3V, 3W and the inductance Lp of the motor windings 7U, 7V, 7W are automatically calculated by another means. Will be described.
In this embodiment, the difference between the ROT signal when the rotor is free-runned at a certain rotational speed and the ROT signal immediately after the drive voltage is supplied to the ROT signal is induced in the motor winding when the rotor is free-runned. Rp and Lp are calculated from the driving voltage and the driving voltage.
FIG. 16 is a block diagram showing the configuration of the control circuit 45 of the sensorless brushless motor according to the eighth embodiment of the present invention. The control circuit 45 is obtained by adding a cable 27 to the control circuit 41 of the fourth embodiment. The cable 27 transmits the output X of the integrator 1 to a microcomputer 30 (not shown) in the drive control circuit.
Since other configurations are the same as those of the control circuit 41, the same components as those of the control circuit 41 are given the same numbers.

Next, the theoretical formula that is the basis of the present embodiment will be described.
First, various quantities are defined as follows.

(Equation 13)
Angular frequency of rotor = ω
Output Iu−Iv = I of the differential amplifier 9
Motor windings 7U and 7V when the motor 5 is driven
Voltage difference Vu−v = Vd
Actual value of induced electromotive force of motor winding when rotor 6 is free running Viu−v = Vir
Estimated value of induced electromotive force of motor winding when rotor 6 is free running Viu−v = Vie
Actual value of Rp = Rpr
Estimated value of Rp = Rpe
Actual value of Lp = Lpr
Estimated value of Lp = Lpe
Phase difference between Vir and Vd = θ1
Phase difference between Vir and Vie = θ2

Here, the estimated value Rpe of Rp and the estimated value Lpe of Lp are assumed values stored in the microcomputer 30. The initial value is Rpe = Lpe = 0, and the measurement of Rpr and Lpr is repeated several times, and is updated sequentially with the actual value of Rpr and the actual value of Lpr each time.
These relationships are represented by a voltage vector diagram as shown in FIG.
Since many high frequency noises are superimposed on the detected voltage value, calculating Rpr and Lpr from the values obtained by integrating the various quantities in FIG. 17 improves the calculation accuracy.
FIG. 18 shows the integration of various quantities shown in FIG. When FIG. 17 is integrated, each vector is rotated by 90 ° counterclockwise, that is, the phase is delayed by 90 ° with respect to the voltage vector. FIG. . The following relational expression holds from the vector diagram of FIG.

(Equation 14)
θ2 = θ1-arcsin (ω × Lpe × I / Vd)
Vie / ω = Vd × cos (θ1−θ2) / ω−RpeI / ω
I × (Rpr−Rpe) / ω = (Vie−Vir * cos θ2) / ω
I × Lpr−I × Lpe = Vir × sin θ2 / ω

  From these relational expressions, the following expressions (14) and (15) are obtained.

(Equation 15)
Rpr = (Vie−Vir × cos θ2) / I + Rpe (14)
Lpr = Vir × sin θ2 / (ω × I) + Lpe (15)

That is, if Vd / ω, Vir / ω, and θ1 are measured, Rpr can be obtained from equation (14) and Lpr can be obtained from equation (15).
The operation of the sensorless brushless motor control circuit 45 will be described below.
As described above, Rpe and Lpe stored in the microcomputer 30 are initialized to Rpe = Lpe = 0.
The rotor 6 is started from a stopped state in an open loop and accelerated to a predetermined rotational speed ω (for example, 20 revolutions per second).
When the rotor rotational speed reaches a predetermined value ω, the current supplied to the motor 5 is stopped for a moment and the rotor 6 is free-runned. At this time, the actual value Vir of the induced electromotive force of the motor winding is detected from the differential amplifier 8. The output of the integrator 1 is stored in the microcomputer 30 via the cable 27.
Next, immediately after measuring Vir, current is supplied to the motor windings 7U, 7V, 7W. At this time, the voltage difference Vd between the motor windings 7U and 7V is measured by the differential amplifier 8. The output of the integrator 1 is stored in the microcomputer 30 via the cable 27. Further, θ1 is acquired simultaneously with the above-described current on / off operation by a method described later.

Next, the measurement of θ1 will be described.
The microcomputer 30 receives the ROT signal as shown in FIG. 19 from the comparator 4 (not shown) in the drive control circuit 19 when the rotor 6 is free running.
The microcomputer 30 has an internal timer (for example, counts up every 10 μs), and counts up the timer pulses while the rotor 6 makes one rotation in synchronization with the ROT signal. For example, if the rotor 6 makes one rotation in 0.1 seconds, the timer count Nr is 1000.
Immediately after this, the microcomputer 30 restarts counting from 0 every time the timer count reaches 1000. Since the rotational speed of the rotor 6 is substantially constant during the free run, the ROT signal and the timer count operation are synchronized.
Next, the microcomputer 30 stores the count number Ne of the counter at the switching point of the ROT signal when the current supply to the motor windings 7U, 7V, and 7W is resumed. As shown in FIG. 19, this ROT signal has substantially the same period as the ROT signal during free run, but the phase advances by θ1.
From the values of Nr and Ne, the microcomputer 30 calculates θ1.
For example, assuming that the count number Nr = 1000 and the count number Ne = 900, the phase shift θ1 between Vir and Vd is obtained as follows.

(Equation 16)
θ1 = (Nr−Ne) / Nr × 360 ° = 36 °

Since Vd / ω, Vir / ω, and θ1 have been obtained as described above, the microcomputer 30 calculates Rpr and Lpr according to the equations (14) and (15).
Next, the microcomputer 30 updates the values of Rpe and Lpe with the values of Rpr and Lpr.
More accurate Rpr and Lpr can be obtained by repeating the above procedure several times when the motor 5 is started.
The control device 45 obtains a magnetic flux estimation signal φu-v based on Rpe and Lpe acquired by the above procedure, and feedback-controls the motor 5. This operation is the same as in the fourth embodiment.
In addition, when the Rpe and Lpe values obtained as a result of the calculation exceed the normal value range (for example, 0.5 [Ω] to 10 [Ω] and 0 [mH] to 1 [mH]), an alarm is output. If a safety device that does not drive the motor 5 is installed, safety can be improved.

In the present embodiment, Rpr and Lpr are automatically measured when the motor 5 is started, and the magnetic flux estimation signal φu-v is obtained based on the measured values. It is not necessary to use and measure.
Therefore, conventionally, when the motor 5 is replaced with another motor at the site where the sensorless brushless motor is used, it has been necessary to remeasure Rpr and Lpr again using a measuring instrument, but the control circuit 45 of the present embodiment. Then, such work becomes unnecessary.
In the present embodiment, Rpe can be measured by Lpe simply by adding the cable 27 from the integrator 1 to the microcomputer 30. Therefore, the control circuit from the sixth embodiment to the seventh embodiment is used. The circuit configuration becomes simpler.
In the present embodiment, Rpr and Lpr are obtained using the motor windings 7U and 7V. However, the present invention is not limited to this, and Rpr and Lpr can be measured by any motor winding.
In the present embodiment, when the start-up time is as long as about 3 to 10 minutes as in the case of a magnetic levitation turbo molecular pump, the acceleration of the rotor is small and θ1 can be measured with high accuracy.

(Ninth embodiment)
In the present embodiment, the control circuit 41 (FIG. 11) of the fourth embodiment is added with a function of temporarily setting the resistance value signal Rp and the inductance value signal Lp to zero, and the like. It is possible to realize the same function as the control circuit 142 (FIG. 3) of the second embodiment using As a result, the control circuit 41 to which the function is newly added can share the function of the control circuit 142, and the operation of the motor 5 can be performed from the low speed rotation where the PLL circuit 16 cannot be locked to the steady rotation. Note that steady rotation means that the rotor is constantly rotated at a predetermined rotation frequency (for example, 30,000 rotations per minute).

  FIG. 21 is a diagram showing a configuration of the control circuit 47 according to the present embodiment. The control circuit 47 can switch the output of the Rp signal setting circuit 14 to either the resistance value signal Rp or zero, and can switch the output of the Lp signal setting circuit 15 to either the inductance value signal Lp or zero. The control circuit 41 is the same as the control circuit 41 (FIG. 11) of the fourth embodiment except for the above points.

The outline of the configuration of the control circuit 47 will be described below.
The motor drive circuit 17 constitutes current supply means for supplying current to the motor 5.
The Rp signal setting circuit 14, the differential amplifier 9, the multiplier 10, and the adder 11 calculate the correction amount of the voltage change due to the combined resistance of the wiring from the power supply device and the motor windings 7U, 7V, and 7W from the line voltage. Resistance correction means for subtracting and correcting this is configured.
The differential amplifier 8 constitutes a line voltage acquisition means for acquiring the line voltage of the motor windings 7U and 7V. The integrator 1 constitutes a magnetic flux signal acquisition means, and integrates the line voltage to obtain a magnetic flux signal (hereinafter referred to as a magnetic flux estimation signal φ because it is a signal for estimating the interlinkage magnetic flux of the motor windings 7U and 7V).
The Lp signal setting circuit 15, the differential amplifier 9, the multiplier 12 and the adder 13 subtract the voltage change due to the reactance of the motor windings 7 U and 7 V from the output of the integrator 1 by subtracting the multiplier from the output of the integrator 1. Reactance correction means for correcting
The comparator 4 constitutes magnetic pole position acquisition means. Since the phase of the magnetic flux estimation signal φ has a corresponding relationship with the magnetic pole position of the rotor 6, the magnetic pole position can be detected by detecting the phase of the magnetic flux signal φ. Therefore, the comparator 4 can detect the position of the magnetic pole by detecting the point where the phase of the magnetic flux estimation signal φ is 2nπ and (2n + 1) π by comparing the ground level with the magnetic flux signal φ. However, n is an integer.

By the way, the Rp signal setting circuit 14 selectively outputs a resistance value signal Rp (= Rc + Ru = Rc + Rw = Rc + Rv) and two values of zero by a signal from the microcomputer 30.
The Lp signal setting circuit 15 is also configured to selectively output two values of an inductance value signal Lp (= Lu = Lw = Lv) and zero by a signal from the microcomputer 30.
When the outputs of the Rp signal setting circuit 14 and the Lp signal setting circuit 15 are both zero, the outputs of the multiplier 12 and the multiplier 10 are both zero, and the differential amplifier 103 and the DC of the control circuit 142 of the second embodiment. A circuit having the same configuration as the circuit including the cutoff filter 102, the integrator 101, and the comparator 104 can be realized. For example, when the control circuit 47 operates in the same manner as the control circuit 142 is the first mode, and the case in which the control circuit 47 operates in the same manner as the control circuit 41 is the second mode, the control circuit 47 has the resistance value signal Rp and The operation is performed in the second mode by outputting the inductance value signal Lp, and the operation is performed in the first mode by setting these signals to zero.

For example, the microcomputer 30 monitors the rotational frequency of the rotor 6 by the ROT signal received from the comparator 4.
Since the ROT signal output from the comparator 4 repeats Hi and Lo every time the rotor 6 rotates in synchronization with the rotation of the rotor 6, for example, by counting the rise or fall of the ROT signal per unit time, the rotor Six rotational frequencies can be calculated.

When the calculated rotation frequency of the rotor 6 is larger than a predetermined value (30 [Hz] during acceleration and 60 [Hz] during deceleration), the microcomputer 30 sets the Rp signal setting circuit 14 and the Lp signal setting circuit. The resistance value signal Rp and the inductance value signal Lp are output from 15 respectively. In this case, since the resistance value signal Rp and the inductance value signal Lp are output, the circuit configuration is the same as that of the control circuit 41.
The microcomputer 30 operates the motor 5 in the same manner as in the fourth embodiment. That is, the comparator 4 generates an ROT signal from the magnetic flux estimation signal φ, and the PLL circuit 16 generates a 12 × f ROT signal from the ROT signal. Then, the microcomputer 30 turns on and off the transistors 21a, 21b, 21c, 21d, 21e, and 21f of the motor driving circuit 17 in synchronization with the 12 × fROT signal, and causes three-phase alternating current to flow through the motor windings 7U, 7V, and 7W. The rotor 6 is rotated (second mode).

The microcomputer 30 sets the outputs of the Rp signal setting circuit 14 and the Lp signal setting circuit 15 to zero when the calculated rotation frequency of the rotor 6 is smaller than a predetermined value. In this case, since the outputs of the multipliers 10 and 12 are zero, the circuit configuration is the same as that of the control circuit 142.
The microcomputer 30 operates the motor 5 in the same manner as in the second embodiment. That is, the transistors 21b, 21c, 21e, and 21f are turned on / off in synchronization with the ROT signal output from the comparator 4 to change the current in the V → W direction (when the drive voltage vector 3 is output) and the W → U direction ( When the drive voltage vector 5 is output), the rotor 6 is rotated alternately (first mode).

As described above, the microcomputer 30 constitutes selection means for selecting the first mode and the second mode according to the rotational frequency of the rotor 6, and when driving the motor 5 in the first mode, A correction invalidating means for invalidating the correction by the adders 11 and 13 by setting the resistance value signal Rp and the inductance value signal Lp to zero is also configured.
The microcomputer 30 also constitutes drive voltage vector drive timing acquisition means. In the first mode, the microcomputer 30 acquires the drive voltage vector drive timing (first drive timing) from the ROT signal of the comparator 4, In the second mode, the drive timing (second drive timing) is acquired from the 12 × fROT signal of the PLL circuit 16. The microcomputer 30 switches to the second drive mode by the first drive voltage vector output means for turning on and off the transistors 21b, 21c, 21e, 21f in the first mode at the first drive timing, and at the second drive timing. 21a, 21b, 21c, 21d, 21e, and 21f are configured as second drive voltage vector output means.

FIG. 22 is a diagram showing the relationship between the operation mode of the control circuit 47 and the rotational frequency of the rotor 6 in this embodiment.
When the control circuit 47 operates in the same manner as the control circuit 142 and the rotor 6 is accelerated or steadily rotated, it is referred to as a two-phase acceleration mode, and conversely, when it is decelerated, it is referred to as a two-phase deceleration mode. If the two are not distinguished, they are simply called the two-phase mode.
Further, when the control circuit 47 operates in the same manner as the control circuit 41 and the rotor 6 is accelerated or steadily rotated, it is called a three-phase acceleration mode, and when it is decelerated, it is called a three-phase deceleration mode. If the two are not distinguished, they are simply called the three-phase mode.
The two-phase deceleration mode and the three-phase deceleration mode are obtained by inverting the field directions of the two-phase acceleration mode and the three-phase acceleration mode, respectively. That is, the field phase is shifted by 180 degrees.
If the polarity of the field is thus reversed, the field generates torque in the direction opposite to the rotation of the rotor 6, as will be described later.

There are eight types of mode switching from (1) to (8) shown in FIG.
The mode switching of (1) is a case where the mode is switched while the rotor 6 is accelerated in the two-phase acceleration mode, and the rotor 6 is decelerated in the two-phase deceleration mode. For example, when the rotation frequency is less than 30 [Hz] after the rotation of the rotor 6 is started, the user stops the motor 5, or the safety device of the motor 5 is operated to stop. There is.
Similarly, the mode switching of (2) is a case of switching from the two-phase deceleration mode to the two-phase acceleration mode. The mode switching of (3) is when switching from the two-phase acceleration mode to the three-phase acceleration mode. (4) Mode switching is when switching from the three-phase acceleration mode to the two-phase deceleration mode. (5) Mode switching is when switching from the two-phase deceleration mode to the three-phase acceleration mode. (6) Mode switching is when switching from the three-phase deceleration mode to the two-phase deceleration mode. (7) Mode switching is when switching from the three-phase acceleration mode to the three-phase deceleration mode. The mode switching of (8) is a case of switching from the three-phase deceleration mode to the three-phase acceleration mode.

  The control circuit 47 according to the present embodiment configured as described above operates as follows. Here, the rotor 6 is rotated in a steady rotation (for example, 30,000 revolutions per minute) through the two-phase acceleration mode and the three-phase acceleration mode from the stop, and then decelerated in the three-phase deceleration mode and the two-phase deceleration mode. The case of stopping will be described. That is, during acceleration, the mode switching of (3) is performed when the rotational frequency of the rotor 6 is 30 [Hz], and mode switching of (6) is performed at 60 [Hz] during deceleration.

[When accelerating from stop to steady rotation]
The control circuit 47 starts the motor 5 in the two-phase acceleration mode. That is, the microcomputer 30 sets the outputs of the Rp signal setting circuit 14 and the Lp signal setting circuit 15 to zero, and then turns on and off the transistors 21b, 21c, 21e, and 21f of the motor drive circuit 17 to drive the drive voltage vector 3. The voltage vector 5 is output alternately. When the transistors 21c and 21f are on, the driving voltage vector 3 is output to the motor windings 7V and 7W. When the transistors 21e and 21b are on, the driving voltage vector 5 is output to the motor windings 7W and 7U. .
FIG. 23 shows the relationship between the drive voltage vector number, the direction of the current flowing through the motor windings 7U, 7V, and 7W and the transistors 21a, 21b, 21c, 21d, 21e, and 21f that are switched. For example, as can be seen from FIG. 23, when the drive voltage vector 1 is output, the direction of the current supplied by the motor drive circuit 17 is from the motor winding 7U to the motor winding 7V, and the transistor to be switched on is the transistor 7a. 7d. Similarly, from FIG. 23, it is possible to read the relationship between other drive voltage vectors, current directions, and transistors that are switched on.

Immediately after the start of the start of the motor 5, the microcomputer 30 alternately outputs the drive voltage vectors 3 and 5 at a low frequency close to DC. Then, the rotor 6 is attracted by the magnetic field and starts rotating. When the rotational frequency of the rotor is about 1 [Hz], the magnetic flux estimation signal φ is output, and the comparator 4 can output the ROT signal.
The microcomputer 30 turns on and off the transistors 21b, 21c, 21e, and 21f in synchronization with the ROT signal received from the comparator 4, and alternately outputs the drive voltage vector 3 and the drive voltage vector 5. When the ROT signal is Hi, the drive voltage vector 5 is output, and when the ROT signal is Lo, the drive voltage vector 3 is output. Thus, when the rotational frequency of the rotor 6 becomes about 1 [Hz], the position of the magnetic pole of the rotor 6 can be detected and the output of the drive voltage vectors 3 and 5 can be feedback-controlled.

The rotation of the rotor 6 is accelerated by the above control. When the microcomputer 30 detects from the ROT signal that the rotational speed of the rotor has reached 30 [Hz], all the transistors of the motor drive circuit 17 are turned off for about 10 [μsec] to 0.1 [sec]. After the drive circuit 17 is paused for a moment, it shifts to the three-phase acceleration mode. This is to prevent the transistors 21a, 21b, 21c, 21d, 21e, and 21f from being short-circuited during mode switching.
When the microcomputer 30 shifts to the three-phase acceleration mode, the microcomputer 30 causes the Rp signal setting circuit 14 and the Lp signal setting circuit 15 to output the resistance value signal Rp and the inductance value signal Lp, respectively, and the transistors 21a, 21b, and 21c of the motor drive circuit 17. , 21d, 21e, 21f are turned on and off, and supply of the drive voltage vectors 1-6 to the motor windings 7U, 7V, 7W is started.
The magnetic flux estimation signal φ is obtained from the adder 13, and thereby the ROT signal is obtained from the comparator 4. The PLL circuit 16 generates a 12 × fROT signal from the ROT signal received from the comparator 4.
The microcomputer 30 receives the 12 × fROT signal from the PLL circuit 13 and feedback-controls switching of the current supplied to the motor windings 7U, 7V, and 7W in the same manner as in the fourth embodiment.
In the above three-phase acceleration mode, the rotor 6 is accelerated to about 30,000 revolutions per minute, and then performs steady rotation.

[When decelerating and stopping from steady rotation]
When the motor 5 is stopped, the magnetic field generated by the motor windings 7U, 7V, and 7W is applied to the magnetic poles of the rotor 6 to apply the brake. First, a method for decelerating the rotor 6 will be described.
FIG. 24 is a view for explaining a mechanism for decelerating the rotor 6 using a field.
FIG. 24A is a diagram showing a case where the rotor 6 is accelerated. The rotor 6 rotates clockwise toward the paper surface. The magnetic poles above and below the rotor 6 schematically represent the magnetic fields generated by the motor windings 7U, 7V, and 7W. In the case of FIG. 22A, it is shown that the field is formed by the motor windings 7U, 7V, and 7W upward in the drawing.
When the magnetic field of the rotor shaft 6 and the magnetic field generated by the motor windings 7U, 7V, 7W are in the relationship shown in the figure, that is, the magnetic pole of the rotor shaft 6 is attracted in the direction in which the rotor shaft 6 rotates. When a strong field is formed, the rotor shaft 6 is accelerated.
On the other hand, as shown in FIG. 24B, the rotor shaft 6 is decelerated when a field is formed that repels the magnetic poles of the rotor shaft 6 in the direction in which the rotor shaft 6 rotates.
From the above consideration, when the rotor shaft 6 is decelerated, the driving voltage vector in the direction opposite to the driving voltage vector output when accelerating is synchronized with the ROT signal in the two-phase decelerating mode, and three-phase decelerating. In the case of the mode, it can be seen that the output should be in synchronization with the 12 × fROT signal.

When stopping the rotor 6 that is rotating in a steady state (about 30,000 revolutions per minute), the microcomputer 30 switches from the three-phase acceleration mode to the three-phase deceleration mode. Based on the 12 × fROT signal output from the PLL circuit 16, a predetermined gate signal is supplied to the transistors 21a, 21b, 21c, 21d, 21e, and 21f, and the currents of the motor windings 7U, 7V, and 7W are sequentially switched. . In this case, the current supplied to the motor windings 7U, 7V, 7W is in the opposite direction to the three-phase acceleration mode. For example, in the case of the three-phase acceleration mode, the transistors 21c and 21b are turned on and the current flows in the V → U direction at the rising edge of the ROT signal that is the starting point of the 12 × f ROT signal from FIG. ). In order for the magnetic field generated by the motor windings 7U and 7V to be in the opposite direction, it is only necessary to pass a current in the U → V direction (drive voltage vector 1). For this purpose, the microcomputer 30 may turn on the transistors 21a and 21d.
In this manner, the microcomputer 30 decelerates the rotor 6 by supplying a current in the opposite direction to that in the three-phase acceleration mode to the motor windings 7U, 7V, 7W in synchronization with the 12 × fROT signal.

The microcomputer 30 monitors the rotational frequency of the rotor 6 based on the ROT signal output from the comparator 4 and detects that the rotational frequency of the rotor 6 has decreased to 60 [Hz]. Then, the microcomputer 30 switches from the three-phase deceleration mode to the two-phase deceleration mode. The control mode of the drive circuit 17 is switched. In order to prevent short-circuiting of the transistors 21a, 21b, 21c, 21d, 21e, and 21f even when the mode is switched, the motor driving circuit 17 is temporarily stopped.
When the microcomputer 30 enters the two-phase deceleration mode, both the outputs of the Rp signal setting circuit 14 and the Lp signal setting circuit 15 are set to zero, and the transistors 21b, 21c, 21e, and 21f are turned on and off to drive voltage vector 3 and driving voltage vector 5 Are output alternately. However, contrary to the two-phase acceleration mode, the driving voltage vector 3 is output when the ROT signal is Hi, and the driving voltage vector 5 is output when the ROT signal is Lo.
As described above, the rotor 6 is quickly stopped by the combined use of the three-phase deceleration mode and the two-phase deceleration mode.

As described above, in the present embodiment, the following effects can be obtained.
In order to realize the function of the control circuit 142 (FIG. 3) of the second embodiment by using a part of the control circuit 41 (FIG. 11) of the fourth embodiment, the rotor 6 Even when the PLL circuit 16 is rotating at a rotation frequency that can be locked (for example, about 20 [Hz]) or less, it is possible to detect the position of the magnetic pole and perform feedback control of the field, and the motor 5 is in steady operation. Even if a sudden load change occurs during operation, operation can be maintained without stepping out.
Therefore, the starting time of the motor 5 can be shortened, and the stability during steady operation can be improved.
In addition, since the single control circuit 47 can control both the startup and the steady operation of the motor 5, it is not necessary to add the control circuit 142 to the control circuit 41, and the manufacturing cost can be reduced.

In the present embodiment, the magnetic flux estimation signal φ output from the integrator 1 is input to the comparator 4 via the adder 13. However, if a DC cutoff filter is inserted between the adder 13 and the comparator 4, For this reason, the operation of the control circuit 47 can be further stabilized.
Since the signal output from the Lp signal setting circuit 15 is zero in the two-phase mode, the output of the multiplier 12 should theoretically be zero. However, since the multiplier 12 is formed by combining various elements such as an operational amplifier, an offset voltage (DC) may be output from the multiplier even if the inductance value signal Lp is zero due to the properties of these elements.
For this reason, in the two-phase mode, a DC component may be superimposed on the magnetic flux estimation signal φ output from the adder 13. On the other hand, since the comparator 4 compares the level of the magnetic flux estimation signal and the ground level, it cannot operate properly when the magnetic flux estimation signal φ is offset. Therefore, by inserting a DC cutoff filter between the adder 13 and the comparator 4 and removing the DC component superimposed by the adder 13, the comparator 4 can be operated more appropriately.
The DC cutoff filter inserted between the adder 13 and the comparator 4 may remain attached even when the control circuit 47 is operating in the three-phase mode. This is because it is desirable that the magnetic flux estimation signal φ input to the comparator 4 is from which the DC component has been removed. Therefore, there is no adverse effect even if a DC cutoff filter is provided on the input side of the comparator 4. This is because if the output of the multiplier 12 has a direct current component due to the characteristics of the element, this can be removed.

(Modification 1 of 9th Embodiment)
FIG. 25 is a diagram showing the configuration of the control circuit 49 according to this modification. In the two-phase mode, the control circuit 49 sets only the output of the Lp signal setting circuit 15 to zero, and outputs the resistance value signal Rp from the Rp signal setting circuit 14 regardless of the mode. Configuration of other circuits Is similar to the control circuit 47. As described above, when the resistance value signal Rp signal is output in the two-phase mode, the characteristic immediately after the start of the control circuit 47 can be improved for the following reason.

The DC cutoff filter 2 (FIG. 21) is configured by combining, for example, a high-pass filter and an integrator, and has normal integration characteristics. Therefore, the response to the input signal is delayed, and even if the input signal changes suddenly, it cannot immediately follow.
Therefore, the DC cutoff filter 2 cannot sufficiently cut off the DC component during a predetermined period (about 1 second) immediately after the control circuit 47 is started, and the DC component is output to the integrator 1.
The adder 11 subtracts the output of the multiplier 10 (DC component of the differential amplifier 8) from the output of the differential amplifier 8. That is, when the output of the multiplier 10 is input to the adder 11, the DC component of the output of the differential amplifier 8 is eliminated.
On the other hand, even when the resistance value signal Rp is output from the Rp signal setting circuit 14 to the multiplier 10 in the two-phase mode, the output of the multiplier 10 is originally for eliminating the DC component of the differential amplifier 8. Therefore, it does not affect the operation of the motor 5 at all. That is, the direct current component is removed from the output of the differential amplifier 8 by the adder 11, and further the direct current component is removed by the direct current cutoff filter 2. Moreover, the direct current component is erased by the adder 11 immediately after the motor 5 is started.

FIG. 26 is a diagram for explaining the difference between the presence / absence of the output of the resistance value signal Rp and the output of the DC cutoff filter 2.
FIG. 26A is a diagram showing a DC component output from the differential amplifier 8, where the vertical axis represents voltage and the horizontal axis represents time. When the control circuit 47 is started at time t1, a signal is output from the differential amplifier 8, and the signal includes a DC component. That is, as shown in the figure, when the control circuit 47 is started, a DC component 81 appears from the differential amplifier 8 in a step shape.
FIG. 26B is a diagram showing the DC component of the signal output from the DC cutoff filter 2 when the output of the Rp signal setting circuit 14 is zero and the output of the multiplier 10 is zero. The axis represents voltage and the horizontal axis represents time. The origin of the time axis is aligned with FIG.
As shown in FIG. 26B, when the output of the multiplier 10 is set to zero, the direct current component 82 is output immediately after the control circuit 47 is started, and is attenuated after a predetermined time.
FIG. 26C is a diagram showing a DC component of a signal output from the DC cutoff filter 2 when the resistance value signal Rp is output from the Rp signal setting circuit 14. The vertical axis represents voltage and the horizontal axis represents time. The origin of the time axis is aligned with FIG.
As shown in FIG. 26 (c), in this case, the DC cutoff filter 2 can cut off the DC component immediately after the control circuit 47 is started.
As described above, by outputting the resistance value signal Rp from the Rp signal setting circuit 14 in the two-phase mode, the output of the DC component from the DC cutoff filter 2 immediately after the start of the control circuit 47 can be suppressed.
Thus, in this modification, the characteristics of the control circuit 47 immediately after starting can be improved.

(Modification 2 of the ninth embodiment)
As described in the first modification of the ninth embodiment, there are eight types of switching between the two-phase mode and the three-phase mode (FIG. 22). Of these, the magnetic flux estimation signal φ output from the adder 13 (FIG. 21) may become unstable when switching from the two-phase mode to the three-phase mode, that is, when switching between the modes (3) and (5). Newly understood.
In this modification, in order to increase the stability of the magnetic flux estimation signal φ, the motor windings 7U, 7V, and 7W do not affect the torque of the rotor 6 when the mode is switched from the two-phase mode to the three-phase mode. Is conducted for a predetermined time (about 1 to 5 seconds).

First, the operation of the magnetic flux estimation signal φ and the like when a minute current is not applied to the motor windings 7U, 7V, and 7W when switching from the two-phase mode to the three-phase mode will be described.
FIG. 27 shows the DC component of the signal of the differential amplifier 8, the magnetic flux estimation signal φ, the current of the W phase when switching from the two-phase deceleration mode to the three-phase acceleration mode after a pause time ((5) mode switching). It is the figure which showed the change of Iw, and the horizontal axis represents time. When the microcomputer 30 switches from the two-phase deceleration mode to the three-phase acceleration mode, the microcomputer 30 has a predetermined pause time between the two-phase mode and the three-phase mode in order to prevent the transistors 21a, 21b, 21c, 21d, 21e, and 21f from being short-circuited. (About 10 μs to 0.1 seconds).
When the microcomputer 30 drives the motor 5 in the three-phase mode after the pause time, a fluctuation (offset) 71 of the DC component of the signal appears in the differential amplifier 8. This variation is caused by variations in the characteristics of the motor 5 and circuit elements.
Since the DC cutoff filter 2 has an integration function, it takes some time to cut the fluctuation 71 of the DC voltage. For this reason, immediately after the fluctuation 71 of the DC voltage appears, it cannot be cut sufficiently, and the integrator 1 integrates the DC component. As a result, the magnetic flux estimation signal φ fluctuates. When the operation in the three-phase mode is performed based on the fluctuation magnetic flux estimation signal φ, the energization phases of the motor windings 7U, 7V, and 7W cannot be switched at the correct timing, and the acceleration operation may not be performed well.

Reference numeral 72 represents the magnetic flux estimation signal φ, and reference numeral 73 represents the envelope of the magnetic flux estimation signal φ. As shown in the figure, when the magnetic flux estimation signal φ is switched to the three-phase acceleration mode, it fluctuates greatly in positive and negative directions, and the fluctuation does not readily attenuate.
On the other hand, current is supplied to the motor windings 7U, 7V, and 7W as follows. Reference numeral 74 represents Iw, and reference numeral 75 represents an envelope of Iw. As shown in the figure, when switching from the two-phase mode to the three-phase mode, the microcomputer 30 generates one from the magnetic flux estimation signal φ after suspending the motor drive circuit 17 with one end Iw (Iu and Iv also) set to zero. The amplitude of Iw is gradually increased in synchronization with the ROT signal, and the mode is shifted to the three-phase mode.
Incidentally, since the ROT signal is generated by the comparator 4 comparing the level of the magnetic flux estimation signal φ and the ground level, if the magnetic flux estimation signal φ fluctuates, the ROT signal does not coincide with the position of the magnetic pole of the rotor 6, and Iw is The correct phase is not energized as shown by the reference numeral 76.
The fluctuation of the magnetic flux estimation signal φ does not converge quickly because the current in the three-phase acceleration mode is supplied to the motor 5 before the fluctuation of the magnetic flux estimation signal φ due to the fluctuation 71 is settled. It seems to cause.

Next, the operation of the magnetic flux estimation signal φ and the like when a small current is applied to the motor windings 7U, 7V, and 7W when switching from the two-phase mode to the three-phase mode will be described.
FIG. 28 is a diagram showing changes in the DC component, the magnetic flux estimation signal φ, and the W-phase current Iw of the differential amplifier 8 in this case.
In this example, when the microcomputer 30 switches from the two-phase deceleration mode (section 51) to the three-phase acceleration mode (section 55), the microcomputer 30 once turns off the current supplied to the motor 5 by the motor drive circuit 17 ( (Section 52, time is about 10 μs to 0.1 seconds), and a three-phase minute current that does not affect the torque of the motor 5 is supplied to the motor 5 for a certain period (about 1 second to 5 seconds) (section 53). ). Thereafter, the current value of the three-phase current is gradually increased (section 54), and the mode is shifted to the three-phase mode (section 55).
In the three-phase acceleration mode, the magnitude of Iw is about 6 [A], and the magnitude of the three-phase minute current is about 0.1 to 0.5 [A].

When the microcomputer 30 starts energizing the three-phase minute current, a fluctuation 58 occurs in the DC component of the signal output from the differential amplifier 8. Then, although the fluctuation of the magnetic flux estimation signal φ occurs (section 57), this quickly converges while the phase minute current is being applied.
The DC component fluctuation 58 occurs when the three-phase mode is started. In this example, since a large three-phase current is not supplied to the motor 5 until the fluctuation of the magnetic flux estimation signal φ converges, the motor windings 7U and 7V are not supplied. Therefore, the fluctuation of the new magnetic flux estimation signal φ due to the current supplied to 7 W does not occur, and as a result, the fluctuation of the magnetic flux estimation signal φ can be quickly converged.

  In the above description, the case where the mode is switched from the two-phase deceleration mode to the three-phase acceleration mode has been described. However, even when the mode is switched from the two-phase acceleration mode to the three-phase acceleration mode, a small 3 By supplying the phase current to the motor 5, the fluctuation of the magnetic flux estimation signal φ can be quickly converged.

  In the present modification described above, the stability of the control circuit 47 when shifting from the two-phase mode to the three-phase mode can be increased, and as a result, the rotor 6 can be normally accelerated.

(Modification 3 of the ninth embodiment)
In this embodiment, the magnetic flux estimation signal φ is passed through the DC cutoff filter before being input to the comparator 4, and the cutoff frequency of the DC cutoff filter is when the rotor 6 is rotating at a low speed and when the rotor 6 is rotating at a high speed. This is what is switched.
As described in the ninth embodiment, even when the output of the Lp signal setting circuit 15 is set to zero in the control circuit 47, the multiplier 12 outputs an offset voltage and does not become zero. As a result, the adder 13 outputs. There is a case where a DC component is superimposed on the magnetic flux estimation signal φ. Therefore, by inserting a DC cutoff filter between the adder 13 and the comparator 4, the DC component superimposed on the magnetic flux estimation signal φ can be removed.
By the way, since the control circuit 47 feedback-controls the motor 5 with the magnetic flux estimation signal φ from the start of the rotor 6 to the steady rotation, the DC cutoff filter has a magnetic flux estimation signal with a small frequency of about 1 [Hz] as the rotational frequency. It is also necessary to pass φ.
On the other hand, the DC cutoff filter is formed using, for example, a high-pass filter. Therefore, the frequency characteristic of the circuit combining the DC cutoff filter 2 and the integrator 1 increases the gain when the cutoff frequency is reduced as will be described later, and a small DC noise may be amplified. When the DC noise is amplified by a circuit combining the DC cutoff filter 2 and the integrator 1, it becomes difficult for the comparator 4 to output the ROT signal correctly.
Therefore, in this modification, a more appropriate magnetic flux is obtained by inserting a DC cutoff filter between the adder 13 and the comparator 4 and switching the cutoff frequency of the DC cutoff filter between the low speed rotation and the high speed rotation of the rotor 6. The estimated signal φ is generated.

The direct current cutoff filter used in this modification is constituted by a high-pass filter. Therefore, the frequency characteristic (gain-frequency characteristic) of the circuit combining the DC cutoff filter and the integrator is obtained by adding the frequency characteristic of the high-pass filter and the frequency characteristic of the integrator.
The frequency characteristics will be described below with reference to FIG.
FIG. 29A is a diagram illustrating frequency characteristics when the cutoff frequencies are f1 and f2 in a high-pass filter having a variable cutoff frequency.
When the cut-off frequency is f1, as shown by the curve 62, the gain increases rapidly from the low frequency side to the frequency f1, and the gain is saturated at a predetermined constant value at a frequency higher than the frequency f1.
Similarly, when the cutoff frequency is f2 (f2> f1), as shown by the curve 61, the gain increases rapidly from the low frequency side to the frequency f2, and at a frequency higher than the frequency f2, the gain is a predetermined constant value. Saturates with value.

FIG. 29B is a diagram showing the frequency characteristics of the integrator. As indicated by line 63, the gain of the integrator decreases linearly as the frequency increases.
FIG. 29C is a diagram illustrating frequency characteristics of a circuit in which a DC cutoff filter and an integrator are combined. Since the DC cutoff filter is formed of the high-pass filter, the gain of the circuit combining the DC cutoff filter and the integrator is the sum of the frequency characteristics of the high-pass filter and the frequency characteristics of the integrator.
The frequency characteristic of the circuit combining the DC cutoff filter and the integrator is such that when the frequency characteristic of the high-pass filter is represented by a curve 61, the gain is maximum at the cutoff frequency f2 as shown by the curve 64, and the frequency characteristic of the high-pass filter is When represented by the curve 62, the gain becomes maximum at the cutoff frequency f1 as shown by the curve 65.
As can be seen from FIG. 29 (c), when the cutoff frequency of the high-pass filter is reduced to reduce the cutoff frequency of the circuit combining the DC cutoff filter and the integrator, the circuit combining the DC cutoff filter and the integrator has a low frequency. , The gain increases by Δ.
Thus, if the cut-off frequency of the circuit combining the DC cut-off filter and the integrator is reduced, the low-frequency magnetic flux estimation signal φ can be passed, but since the gain is increased, DC noise (DC component of noise) is reduced. Will also amplify.
If the cut-off frequency of the circuit combining the DC cut-off filter and the integrator is reduced, the low-frequency magnetic flux estimation signal φ can be passed, so that the operation at the start of the motor is stabilized, but the DC noise is amplified. The operation may become unstable when the motor is in steady operation. On the other hand, if the cutoff frequency of the circuit combining the DC cutoff filter and integrator is increased, the DC noise will not be amplified and the steady operation of the motor will be stable, but the low frequency magnetic flux estimation signal φ will be difficult to pass. May become unstable.
Therefore, it is desirable to change the cutoff frequency of the high-pass filter when the motor 5 is started (it is necessary to pass a low-frequency magnetic flux estimation signal φ) and when the rotational speed of the rotor 6 increases to some extent.

FIG. 30 is a diagram showing a configuration of the control circuit 48 according to this modification. The control circuit 48 newly adds a DC cutoff filter 28 between the adder 13 and the comparator 4 to the control circuit 47 of the ninth embodiment, and further, a cutoff frequency change signal is sent from the microcomputer 30 to the DC cutoff filter 28. And is configured to be transmitted to the DC cutoff filter 2. Constituent elements corresponding to the control circuit 47 are denoted by the same reference numerals. In the present embodiment, the cutoff frequency of the DC cutoff filter 2 as well as the DC cutoff filter 28 can be changed. This is to prevent the DC component from being integrated by the integrator 1 by increasing the cutoff frequency when the rotational speed of the rotor 6 increases to some extent. Only the DC cutoff filter 28 may be configured to change the cutoff frequency.
The DC cutoff filters 28 and 2 are both high-pass filters. Since it is not preferable that a DC component is input to the integrator 1 for integration, the cutoff frequency of the DC cutoff filter 2 is made variable in this modification.
The microcomputer 30 sets the cut-off frequency of the DC cut-off filters 28 and 2 to f1 [Hz] for a predetermined time (for example, 10 seconds) after starting the motor 5 from when the motor 5 is stopped, and sets the cut-off frequency after the predetermined time has elapsed. f2 (f1 <f2) [Hz] is set higher.
The microcomputer 30 sets these cutoff frequencies to f1 [Hz] by setting the cutoff frequency change signal to the reset side and sending them to the DC cutoff filters 28 and 2 and sends the cutoff frequency change signal to the set side. Thus, these cutoff frequencies can be set to f2. That is, the microcomputer 30 constitutes switching means for switching the cutoff frequency.
In this modification, f1 = 0.05 [Hz] and f2 = 0.5 [Hz].
Since the gain is proportional to the frequency, the gain at the cutoff frequency f1 is 10 times the gain at the cutoff frequency f2.

The control circuit 48 configured as described above operates as follows.
When the motor 5 is started from a stopped state, the microcomputer 30 outputs a cutoff frequency change signal to the DC cutoff filters 28 and 2 on the reset side, and sets these cutoff frequencies to f1 = 0.05 [Hz]. .
Thereafter, when the microcomputer 30 starts the motor drive circuit 17 in the two-phase acceleration mode, the microcomputer 30 starts measuring the elapsed time since starting the motor 5 at the same time.
When 10 seconds have elapsed after the motor 5 is started, the microcomputer 30 sets the cutoff frequency change signal to the set side and outputs it to the DC cutoff filters 28 and 2, and sets these cutoff frequencies to f2 = 0.5 [Hz]. Set.
Thereafter, the motor 5 is operated in the same manner as in the ninth embodiment.

  In this modification described above, the cutoff frequency of the DC cutoff filters 28 and 2 is reduced (eg, 0.05 [Hz]) only for a short time (eg, 10 seconds) after the start of rotation, and the cutoff frequency is reached after the short time has elapsed. Since the motor 5 is operated at a high value (for example, 0.5 [Hz]), signals (magnetic flux estimation signal φ, ROT signal, etc.) for estimating the magnetic pole position at the start of the motor 5 and steady rotation are stable, Improves motor stability.

  In this modification, the microcomputer 30 detects whether the rotational frequency of the rotor 6 is larger or smaller than a predetermined value from the ROT signal. If the rotational frequency is equal to or lower than the predetermined value, the DC cutoff filter 28, When the cutoff frequency of 2 is set to f1 [Hz] and the rotational frequency is higher than a predetermined value, the cutoff frequency can be set to f2 (f1 <f2) [Hz].

  The ninth embodiment and the first to third modifications of the ninth embodiment have been described above. However, this can be performed individually or in any combination. You can also.

(Tenth embodiment)
In this embodiment, a vacuum pump that controls a motor by the control circuit 47 described in the ninth embodiment will be described.
In this embodiment, a magnetic bearing type turbo molecular pump will be described as an example of a vacuum pump.
FIG. 31 is a diagram showing an example of a sectional view in the axial direction of the rotor shaft 303 of the turbo molecular pump 301.
The casing 316 has a cylindrical shape and forms an exterior body of the turbo molecular pump 301.

A rotor shaft 303 is installed at the center of the casing 316.
Magnetic bearing portions 308, 312, and 320 are provided at the upper, lower, and bottom portions of the rotor shaft 303 toward the paper surface, respectively. When the turbo molecular pump 301 is in operation, the rotor shaft 303 is magnetically levitated in the radial direction (the radial direction of the rotor shaft 303) by the magnetic bearing portions 308 and 312 and supported in a non-contact manner, and is thrust by the magnetic bearing portion 320. Magnetically levitated in the direction (axial direction of the rotor shaft 303) and supported in a non-contact manner.
These magnetic bearing portions constitute a so-called five-axis control type magnetic bearing, and the rotor shaft 303 and the rotor 311 fixed to the rotor shaft 303 can rotate around the axis of the rotor shaft 303. .

In the magnetic bearing portion 308, four electromagnets are arranged around the rotor shaft 303 so as to face each other at 90 °. The rotor shaft 303 is formed of a high permeability material such as iron and is attracted by the magnetic force of these electromagnets.
The displacement sensor 309 is a radial sensor that detects the displacement of the rotor shaft 303 in the radial direction. When the control device 325 detects that the rotor shaft 303 is displaced from the predetermined position in the radial direction by the displacement signal from the displacement sensor 309, the control device 325 adjusts the magnetic force of each electromagnet so as to return the rotor shaft 303 to the predetermined position. Operate. The adjustment of the magnetic force of the electromagnet is performed by feedback controlling the excitation current of each electromagnet.

  In this manner, the control device 325 feedback-controls the magnetic bearing portion 308 based on the signal from the displacement sensor 309, whereby the rotor shaft 303 is magnetically levitated in the radial direction with a predetermined clearance from the electromagnet in the magnetic bearing portion 308. And is held in the space without contact.

The configuration and operation of the magnetic bearing portion 312 are the same as those of the magnetic bearing portion 308.
In the magnetic bearing portion 312, four electromagnets are arranged around the rotor shaft 303 every 90 °, and the rotor shaft 303 is non-radially arranged in the radial direction by the magnetic bearing portion 312 due to the attractive force of the magnetic force of these electromagnets. Held in contact.
The displacement sensor 313 is a radial sensor that detects the displacement of the rotor shaft 303 in the radial direction.

When the rotor shaft 303 receives a radial displacement signal from the displacement sensor 313 from the displacement sensor 313, the control device 325 feedback-controls the excitation current of the electromagnet so that the displacement is corrected and the rotor shaft 303 is held at a predetermined position.
The control device 325 feedback-controls the magnetic bearing portion 312 based on the signal of the displacement sensor 313, whereby the rotor shaft 303 is magnetically levitated in the radial direction by the magnetic bearing portion 312 and is held in a non-contact manner in the space. .

The magnetic bearing portion 320 provided at the lower end of the rotor shaft 303 includes a disk-shaped metal disk 318, electromagnets 314 and 315, and a displacement sensor 317, and holds the rotor shaft 303 in the thrust direction.
The metal disk 318 is made of a high magnetic permeability material such as iron, and is fixed perpendicularly to the rotor shaft 303 at the center thereof. An electromagnet 314 is installed on the metal disk 318 and an electromagnet 315 is installed below. The electromagnet 314 attracts the metal disk 318 upward by magnetic force, and the electromagnet 315 attracts the metal disk 318 downward. The control device 325 appropriately adjusts the magnetic force exerted by the electromagnets 314 and 315 on the metal disk 318 so that the rotor shaft 303 is magnetically levitated in the thrust direction and held in the space without contact.

The displacement sensor 317 is an axial sensor that detects the displacement of the rotor shaft 303 in the thrust direction, and transmits the detected signal to the control device 325. The control device 325 detects the displacement of the rotor shaft 303 in the thrust direction based on the displacement detection signal received from the displacement sensor 317.
When the rotor shaft 303 moves in one of the thrust directions and is displaced from a predetermined position, the control device 325 adjusts the magnetic force by feedback controlling the excitation current of the electromagnets 314 and 315 so as to correct this displacement, It operates to return the rotor shaft 303 to a predetermined position. By this feedback control of the control device 325, the rotor shaft 303 is magnetically levitated and held at a predetermined position in the thrust direction.
As described above, the rotor shaft 303 is held in the radial direction by the magnetic bearing portions 308 and 312 and is held in the thrust direction by the magnetic bearing portion 320. Therefore, the rotor shaft 303 is non-rotated around the axis due to magnetic levitation. It is supported by contact.

In the axial direction of the rotor shaft 303, a protective bearing 306 is provided above the magnetic bearing portion 308, and a protective bearing 307 is provided below the magnetic bearing portion 312.
The rotor shaft 303 is magnetically levitated by the magnetic bearing portions 308, 312, and 320 and is held in a non-contact manner in the space. However, the rotor shaft 303 is held due to vibration around the axis of the rotor shaft 303. There may be a large deviation from the position. In such a case, the protective bearings 306 and 307 are provided to prevent the rotor shaft 303 from contacting the electromagnets of the magnetic bearing portions 308, 312, and 320 and the motor portion 310 from contacting the permanent magnets. Yes.
When the rotor shaft 303 moves a certain amount or more from a predetermined position, the rotor shaft 303 contacts the protective bearings 306 and 307, and the movement of the rotor shaft 303 is physically limited.

  The rotor shaft 303 is provided with a motor portion 310 between the magnetic bearing portions 308 and 312. The motor unit 310 is composed of a DC brushless motor, and will be described in detail later with reference to FIG. The motor unit 310 generates torque and rotates the rotor shaft 303.

The rotor 311 is fixed to the rotor shaft 303 by bolts 305. When the rotor shaft 303 is driven and rotated by the motor unit 310, the rotor 311 also rotates accordingly.
The rotor blades 321 are attached to the rotor 311 on the air inlet 324 side by a predetermined angle from the plane perpendicular to the axis of the rotor shaft 303 and radially attached to the rotor 311. The rotor blades 321 are fixed to the rotor 311 and rotate at a high speed together with the rotor 311.
In addition, the stator blades 322 are fixed to the casing 316 alternately with the stage of the rotor blades 321 toward the inside of the casing 316. The stator blade 322 is fixed to the casing 316 at a predetermined angle from a plane perpendicular to the axis of the rotor shaft 303.

The portion of the rotor 311 on the exhaust port 319 side has a cylindrical outer peripheral surface. A cylindrical thread groove spacer 302 is disposed on the outer periphery of the rotor 311 with a predetermined gap from the outer peripheral surface. The thread groove spacer 302 is made of, for example, aluminum. A thread groove pump part is formed by the thread groove spacer 302 and the rotor 311.
A spiral thread groove 304 is formed on the inner peripheral surface of the thread groove spacer, and the depth of the thread groove 304 decreases as it goes down. When the rotor 311 rotates, the gas is transported down the screw groove 304, but the depth of the screw groove 304 decreases as it goes down, so that the gas is compressed as it is transported through the screw groove 304. It has become.

The control device 325 is connected to the connector of the turbo molecular pump 1 and controls the magnetic bearing units 308, 312, 320 and the motor unit 310.
The control device 325 stores the control circuit 47 described in the ninth embodiment, and the control circuit 47 controls the motor unit 310.
The gas sucked from the intake port 324 is compressed by the action of the rotor blades 321 and the stator blades 322 and sent to the thread groove pump unit.
The gas sent to the thread groove pump section is transported to the lower stage while being guided through the thread groove 304, further compressed, and then discharged from the exhaust port 319.

FIG. 32 is a schematic diagram showing a cross section (FIG. 31) of the motor unit 310 in the XX direction. The motor unit 310 is a part corresponding to the motor 5 of the control circuit 47, and the motor unit 310 is composed of a rotor shaft 303 to which a permanent magnet is fixed and an electromagnet (stator coil) disposed around the rotor shaft 303. It is composed of a mold motor.
The motor unit 10 includes U-phase electromagnets 326e and 326f, V-phase electromagnets 326c and 326d, and W-phase electromagnets 326a and 326b. These electromagnets are arranged concentrically every 60 degrees and in-phase electromagnets are opposed to each other, and motor windings 7U, 7V, and 7W corresponding to the respective phases are wound around each electromagnet. The cores of these electromagnets are composed of laminated steel plates and the like, and are excited when current is supplied to the motor windings. Further, for example, when a current is passed through the motor winding 7U, the U-phase electromagnet 326e becomes the N pole and the U-phase electromagnet 326f becomes the S pole, so that the motor windings 7U, 7V, 7W It is wound so that the polarity of is reversed.

  The example of FIG. 32 shows that the drive voltage vector 1 is output (current flows from the motor winding 7U to 7V), and the electromagnets 326c and 326e are S poles. When the output drive voltage vector changes from 2 → 3 → 4 → 5 → 6, the S pole electromagnet changes from 326e, 326b → 326b, 326d → 326d, 326f → 326f, 326a → 326a, 326c (these When the driving voltage vector makes one round from 1 to 6, the magnetic field generated on the rotor shaft 303 makes one round in the rotation direction of the rotor shaft 303.

On the other hand, two permanent magnets 328 and 239 are fixed on the rotor shaft 303, and the surface of each electromagnet facing the electromagnet has N poles (permanent magnets 328) for each half circumference of the rotor shaft 303 in the circumferential direction. And an S pole (permanent magnet 329).
As shown in FIG. 32, when the drive voltage vector 1 is output, the electromagnets 326c and 326e become the S pole, the electromagnets 326d and 326f become the N pole, the permanent magnet 328 is located on the upper side of the page, and the 329 is located on the lower side. In this case, the permanent magnet 328 is attracted by the electromagnets 326c and 326e, and the permanent magnet 329 is attracted by the electromagnets 326d and 326f, so that the rotor shaft 303 generates a clockwise torque toward the paper surface.
In this manner, the rotor shaft 303 is rotated by sequentially outputting the drive voltage vector 1 → 2 → 3 → 4 → 5 → 6 so as to generate torque on the rotor shaft 303 while detecting the positions of the magnetic poles 328 and 329. Can be made. The positions of the magnetic poles 328 and 329 are detected by the magnetic flux estimation signal φ.
Conversely, when the permanent magnets 328 and 329 are at the positions shown in FIG. 32, the electromagnets 326c and 326e are set to the N pole and the electromagnets 326d and 326f are set to the S pole (that is, the polarity is reversed). Thus, a counterclockwise torque is generated, and the rotor shaft 303 (assuming rotating clockwise) can be braked.

The turbo molecular pump 301 configured as described above operates as follows.
When starting from the stop state, the control device 325 drives the magnetic bearing portions 308, 312, and 320 to magnetically float the rotor shaft 303, and then drives the motor portion 310 and the motor portion 310 in the two-phase acceleration mode. The shaft 303 is rotated.
When the rotation frequency of the rotor shaft 303 reaches a frequency (for example, 30 Hz) at which the PLL circuit can be locked, the control device 325 switches the drive of the motor unit 310 to the three-phase acceleration mode and performs steady rotation (for example, 30,000 rotations per minute). Until the rotor shaft 303 is accelerated. Then, the rotation of the rotor shaft 303 is maintained as it is in the three-phase acceleration mode.
When the rotor shaft 303 rotates, the gas in the chamber (exhaust vessel) to which the turbo molecular pump 301 is connected is sucked from the intake port 324 and compressed by the action of the rotor blade 321 and the stator blade 322.
The gas compressed by the rotor blades 321 and the stator blades 322 is further compressed while being transported through the thread groove 304 of the thread groove pump portion, and then discharged from the exhaust port 319.
When the turbo molecular pump 301 is stopped from the steady operation state, the control device 325 decelerates the rotation of the rotor shaft 303 to a predetermined rotation frequency (for example, about 60 [Hz]) in the three-phase deceleration mode, and then performs the two-phase operation. Switch to deceleration mode and further decelerate to stop. The control device 325 stops the magnetic bearing portions 308, 312, and 320 after stopping the rotation of the rotor shaft 303.
Although the turbo molecular pump 301 is operated in the order of the two-phase acceleration mode → the three-phase acceleration mode → the three-phase deceleration mode → the two-phase deceleration mode as described above, the mode switching is as described in the ninth embodiment. There are 8 types.

As described above, in the present embodiment, the following effects can be obtained.
In the two-phase acceleration mode, the rotation of the rotor shaft 3 can be started regardless of the initial positions of the magnetic poles 328 and 329. Therefore, it is not necessary to DC brake the magnetic poles 328 and 329 when the rotor shaft 3 is started.
Even at a rotation frequency at which the PLL circuit 16 cannot be locked, the magnetic pole positions 328 and 329 can be detected by the magnetic flux estimation signal φ and the field can be feedback-controlled.
With the above two points, the startup time of the turbo molecular pump 301 can be shortened, and startup failure can be suppressed.
Further, even when the turbo molecular pump 301 is in operation and there is a disturbance such as an outside air entry and the rotation frequency of the rotor shaft 3 changes greatly, the rotor shaft 3 is controlled without being stepped out by the magnetic flux estimation signal φ. Can do.

As described above, in the present embodiment, the motor unit 310 is composed of two permanent magnets fixed to the rotor shaft 3 and six electromagnets disposed around the permanent magnet (three-phase two-pole), but is not limited thereto. However, it may be composed of other numbers of permanent magnets or electromagnets.
In addition, the vibration of the rotor shaft 303 when the permanent magnets 328 and 329 are DC-brake at the time of activation may be attenuated by mechanical friction using the protective bearings 306 and 307 without using the magnetic bearing portions 308, 312, and 320. Although it is conceivable, it is necessary to configure a circuit for DC braking by stopping the magnetic bearing portions 308, 312, 320, and the protective bearings 306, 307 are worn, which is not preferable.
In the present embodiment, the control device 325 is equipped with the control circuit 47 of the ninth embodiment as the control circuit of the motor unit 310, but the present invention is not limited to this, and the first embodiment to the first embodiment. A control device (control circuit) according to each modification of the eighth embodiment and the ninth embodiment can also be provided. When the control circuit 143 according to the third embodiment is used, the rotational speed sensor 125 is attached in the vicinity of the rotor shaft 103. For example, a permanent magnet can be attached as a target to the lower end of the rotor shaft 303 and detected by a hall sensor or the like.
In this embodiment, an example of a magnetic bearing type turbo molecular pump is given, but the method of the bearing is not limited to this, and a mechanical type bearing such as a rolling bearing or a sliding bearing may be used. . As the slide bearing, a hydrostatic bearing or a hydrodynamic bearing using gas or liquid may be used.

(Eleventh embodiment)
In the present embodiment, the motor 5 is constituted by an outer rotor type motor.
An example of the configuration of the outer rotor type motor 5 will be described below with reference to FIG. The description of the configuration and operation of the control circuit 47 is the same as in the ninth embodiment, and will not be repeated.
The rotor 6 includes permanent magnets 86 and 87, a yoke 88, a rotor shaft (not shown), and the like.
The yoke 88 is iron or the like formed in a cylindrical shape, and permanent magnets 86 and 87 are fixed to the inner peripheral surface. In the present embodiment, the permanent magnet has two poles, and on the inner peripheral surface side, the permanent magnet 86 has an S pole and the permanent magnet 87 has an N pole.
On the other hand, the stator includes a stator core 85, motor windings 7U, 7V, 7W, and the like. The stator core 85 has U-phase, V-phase, and W-phase magnetic poles formed every 120 degrees, and motor windings 7U, 7V, and 7W are wound around the magnetic poles.

The motor 5 configured as described above operates as follows.
[In 2-phase mode]
The microcomputer 30 (control circuit 47, see FIG. 11) turns on and off the transistors 7b, 7c, 7e and 7f of the motor drive circuit 17 in synchronization with the ROT signal output from the comparator 4, and drives the drive voltage vectors 3, 5 Are alternately output to the motor windings 7U, 7V, and 7W. As a result, the rotor 6 rotates.
When the drive voltage vector 3 is output, the current flows in the V → W direction, and when the drive voltage vector 5 is output, the current flows in the W → U direction.
The microcomputer 30 sets the output of the multiplier 12 and the multiplier 10 to zero by setting the inductance value signal Lp and the resistance value signal Rp to zero.
[Three-phase mode]
The microcomputer 30 turns on and off the transistors 7a, 7b, 7c, 7d, 7e, and 7f of the motor drive circuit 17 in synchronization with the 12 × fROT signal output from the PLL circuit 16, and drives the drive voltage vector 1 to the drive voltage vector 6 Are sequentially output to the motor windings 7U, 7V, and 7W. Thereby, the rotor 6 rotates.
The microcomputer 30 causes the Rp signal setting circuit 14 and the Lp signal setting circuit to output a resistance value signal Rp and an inductance value signal Lp, respectively.

In the above example, the motor 5 is driven by the control circuit 47 of the ninth embodiment. However, the control according to each modification of the first to eighth embodiments and the ninth embodiment. It can also be driven using a device. Among these, when driven by the control circuit 143 according to the third embodiment, the rotational speed sensor 125 (FIG. 5) is installed in the motor 5.
Further, the motor 5 can be a motor of another form, for example, a motor in which the number of poles of the stator coil is 6 and the number of poles of the rotor 6 is 6.
Further, the motor 5 of the present embodiment is a radial air gap type motor having an air gap in the radial direction, but may be an axial air gap type motor having an air gap in the axial direction (direction of the rotation shaft). .

It is the block diagram which showed the control circuit of the brushless motor which concerns on the 1st Embodiment of this invention. It is a wave form diagram of the control circuit of the brushless motor which concerns on the 1st Embodiment of this invention. It is the block diagram which showed the control circuit of the brushless motor which concerns on the 2nd Embodiment of this invention. It is a wave form diagram of the control circuit of the brushless motor which concerns on the 2nd Embodiment of this invention. It is the block diagram which showed the control circuit of the brushless motor which concerns on the 3rd Embodiment of this invention. It is a wave form diagram of the control circuit of the brushless motor which concerns on the 3rd Embodiment of this invention. 4 is a table showing the relationship between a phase difference Y, a drive voltage vector, and a phase delay amount D. It is the figure which showed the structure of the brushless motor. It is the figure which showed the drive voltage vector. It is the figure which showed the resistance value of the motor winding, the inductance, and the resistance value of a connection cable. It is the block diagram which showed the structure of the control circuit of 4th Embodiment. It is a wave form diagram showing waveforms, such as a current of a motor winding at the time of rotating a rotor using the circuit, a voltage, magnetic flux presumption signal φu-v, and a ROT signal. It is the block diagram which showed the structure of the control circuit of 5th Embodiment. It is the block diagram which showed the structure of the control circuit of 6th Embodiment. It is the block diagram which showed the structure of the control circuit of 7th Embodiment. It is the block diagram which showed the structure of the control circuit of 8th Embodiment. It is the vector figure which showed the relationship of the motor voltage etc. of 8th Embodiment. FIG. 18 is a vector diagram showing the relationship of vectors obtained by integrating the vectors in FIG. 17. It is the figure which showed the shift | offset | difference of the ROT signal when the rotor is free-runned in the eighth embodiment and when the drive current is supplied. It is the figure which showed the timing of the magnetic pole position detection of the rotor of the conventional sensorless brushless motor. It is the figure which showed the structure of the control circuit which concerns on 9th Embodiment. It is the figure which showed the relationship between the operation mode of the control circuit in this Embodiment, and the rotational frequency of a rotor. It is the figure which showed the relationship between the number of a drive voltage vector, the direction of the electric current sent through a motor winding, and the transistor to switch. It is a figure for demonstrating the mechanism which decelerates a rotor with a magnetic field. It is the figure which showed the structure of the control circuit which concerns on the modification of 9th Embodiment. (A) is the figure which showed the direct current component output from a differential amplifier, (b) is the figure showing the direct current component output from the direct current cutoff filter when the output of a multiplier is made into zero. (C) is a diagram showing a DC component output from the DC cutoff filter when a predetermined signal is output from the multiplier. It is the figure which showed the direct current component of the signal of a differential amplifier, the magnetic flux estimation signal, and the change of the electric current of a W phase at the time of switching from a two-phase deceleration mode to a three-phase acceleration mode through a rest time. It is the figure which showed the change of the differential amplifier, the magnetic flux estimation signal, and the electric current of W phase at the time of energizing a small current to a motor winding, when changing from 2 phase mode to 3 phase mode. (A) is the figure which showed the frequency characteristic at the time of making cutoff frequency f1 and f2 with the high pass filter with which cutoff frequency is variable, (b) is the figure which showed the frequency characteristic of the integrator, (C) is the figure which showed the frequency characteristic of the circuit which combined the direct current | flow cutoff filter and the integrator. It is the figure which showed the structure of the control circuit which concerns on the modification 3 of 9th Embodiment. It is the figure which showed an example of sectional drawing of a turbo-molecular pump. It is the schematic diagram which showed the cross section of the motor part. It is a figure showing an example of composition of an outer rotor type motor.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Integrator 2 DC interruption | blocking filter 3U, 3V, 3W Connection cable 4 Comparator 5 Motor 6 Rotor 7U, 7V, 7W Motor winding 8 Differential amplifier 9 Differential amplifier 10 Multiplier 11 Adder 12 Multiplier 13 Adder 16 PLL Circuit 17 Motor drive circuit 19 Drive control circuit 22 Low-pass filter 23 Low-pass filter 24 Low-pass filter 25 Low-pass filter 26 Sampling circuit 28 DC cutoff filter 30 Microcomputer 101 Integrator 102 DC cutoff filter 103 Differential amplifier 104 Comparator 107U Motor winding 107V Motor Winding 107W Motor winding 108U Resistor 108V Resistor 108W Resistor 112 Rotor 115 Motor drive circuit 125 Rotational speed sensor 126 Rotational speed detection circuit 127 PLL Path 128 Storage section 130 Microcomputer 301 Turbo molecular pump 302 Screw groove spacer 303 Rotor shaft 305 Bolt 306 Protective bearing 307 Protective bearing 308 Magnetic bearing section 309 Displacement sensor 310 Motor section 312 Magnetic bearing section 313 Displacement sensor 314 Electromagnet 315 Electromagnet 316 Casing 317 Displacement sensor 319 Exhaust port 320 Magnetic bearing portion 321 Rotor blade 322 Stator blade 325 Controller

Claims (14)

A rotor with magnetic poles;
A plurality of motor windings for rotating the rotor;
Current supply means for supplying current to the plurality of motor windings;
Magnetic flux acquisition means for acquiring an interlinkage magnetic flux of at least one of the motor windings by the magnetic pole;
Magnetic pole position acquisition means for acquiring the position of the magnetic pole from a change in the interlinkage magnetic flux acquired by the magnetic flux acquisition means,
The magnetic flux acquisition means includes a voltage processing calculation means for performing a voltage processing calculation for subtracting a voltage based on a voltage drop due to at least a resistance from a line voltage of the motor winding, and the interlinkage when performing the voltage processing calculation. and a DC blocking unit for removing a DC component superimposed from a signal for calculating a magnetic flux,
The control circuit for a sensorless brushless motor, wherein the current supply means switches a current of the motor winding based on the magnetic pole position acquired by the magnetic pole position acquisition means.
The magnetic flux acquisition means includes
First acquisition means for acquiring a line voltage value between two predetermined motor windings;
Second acquisition means for acquiring a voltage drop due to a combined resistance of a resistance of two predetermined motor windings , a power supply device constituting the current supply means, and a resistance of a cable connecting the motor windings;
Third acquisition means for acquiring a value obtained by multiplying a difference between currents of two predetermined motor windings by an inductance value of the two predetermined motor windings;
An integral value acquisition means for integrating after subtracting the value acquired by the second acquisition means from the value acquired by the first acquisition means;
The sensorless brushless motor control circuit according to claim 1, further comprising: a subtracting unit that subtracts the value acquired by the third acquiring unit from the value acquired by the integrated value acquiring unit.
The value of the combined resistance is
DC current supply means for supplying a DC current to the two predetermined motor windings;
It is acquired by a combined resistance value acquisition means comprising a first calculation means for calculating the value of the combined resistance by dividing the line voltage value by the current value of the direct current. Item 3. A sensorless brushless motor control circuit according to Item 2.
The inductance is
High-frequency current supply means for supplying a high-frequency current to the two predetermined motor windings;
A line voltage value acquiring means for acquiring a line voltage value of the two motor windings when the high frequency current is supplied;
The line voltage value is obtained by an inductance obtaining means comprising: a current value of the high-frequency current; and a second computing means for obtaining a value obtained by dividing the line frequency value by the frequency of the high-frequency current and a predetermined constant. The sensorless brushless motor control circuit according to claim 2, wherein:
The inductance is
Rotor rotating means for rotating the rotor by switching the current of the motor winding in an open loop;
Sampling means for sampling the integrated value acquired by the integrated value acquiring means before and after switching the current of the motor winding;
Current peak value acquisition means for acquiring a peak value of a current value supplied to two predetermined motor windings;
Inductance acquisition comprising: third arithmetic means for dividing the absolute value of the difference between the integrated values obtained before and after the current switching obtained by the sampling means by the current peak value obtained by the current peak value obtaining means The sensorless brushless motor control circuit according to claim 2, wherein the control circuit is acquired by means.
Assumed magnetic flux acquisition means for acquiring interlinkage magnetic flux between two predetermined motor windings using the assumed value of the combined resistance and the assumed value of the inductance;
Line voltage values of two predetermined motor windings when the rotor is rotated at a predetermined angular velocity by the rotor rotating means;
A line voltage between two predetermined motor windings when the current of the motor winding is stopped and the rotor is free-runned at the predetermined angular velocity;
A signal obtained by the assumed magnetic flux acquisition means when the supply of the current is stopped;
The phase difference of the signal obtained by the assumed magnetic flux acquisition means when the supply of the current is resumed,
3. The sensorless brushless motor control circuit according to claim 2, further comprising correction means for correcting the assumed value of the combined resistance and the assumed value of the inductance.
Among a plurality of motor windings for rotating a rotor provided with magnetic poles, a voltage difference between two predetermined phases having the same phase and magnitude as the voltage drop due to the inductance of the motor winding is at least a voltage drop due to resistance. and a subtraction of the voltage based, the processing consisting of integration, and the magnetic flux signal acquiring means for acquiring a magnetic flux signal,
When performing the arithmetic processing, a DC blocking unit that removes a DC component superimposed on the voltage difference signal; and
From the arithmetic processing magnetic flux signals on the basis of a signal from the removed DC component superimposed by the DC blocking portion, the first driving for obtaining the drive timing of the part of the drive voltage vector of the output possible driving voltage vector Timing acquisition means;
First drive voltage vector output means for outputting the partial drive voltage vector in synchronization with the drive timing acquired by the first drive timing acquisition means;
Second drive timing acquisition means for acquiring the drive timing of the output drive voltage vector by multiplying the timing obtained from the magnetic flux signal acquired by the magnetic flux signal acquisition means;
Second drive voltage vector output means for outputting the output drive voltage vector in synchronization with the drive timing acquired by the second drive timing acquisition means;
Selecting means for selecting the first drive voltage vector output means and the second drive voltage vector output means;
A control circuit for a sensorless brushless motor, comprising:
Current supply means for supplying current to a plurality of motor windings for rotating a rotor having magnetic poles;
Among the plurality of motor windings, a line voltage acquisition means for acquiring a line voltage between two predetermined motor windings having the same phase and magnitude as the voltage drop due to the inductance of the motor windings;
From the line voltage acquired by the line voltage acquisition means, the resistance of the predetermined two motor windings, and the resistance of the connection cable connecting the motor winding and the power supply device constituting the current supply means Resistance correction means for correcting a change in voltage due to the combined resistance;
Magnetic flux signal acquisition means for integrating the line voltage corrected by the resistance correction means to acquire a magnetic flux signal;
Reactance correction means for correcting a change due to reactance of the two predetermined motor windings among the magnetic flux signals acquired by the magnetic flux signal acquisition means;
Magnetic pole position acquisition means for acquiring the position of the magnetic pole from the magnetic flux signal corrected by the reactance component correction means;
Among the resistance component correction unit and the reactance component correction unit, the correction component invalidation unit disables at least the reactance component correction unit and disables the correction.
When the rotation speed of the rotor is equal to or less than a predetermined rotation,
The correction invalidation means invalidates at least the reactance correction means, and the current supply means switches a current to flow through the predetermined two motor windings based on the magnetic pole position acquired by the magnetic pole position acquisition means. Supplying current to the two motor windings in a first mode;
A second mode for switching the current of the motor winding based on the magnetic pole position acquired by the magnetic pole position acquisition means without using the correction invalidation means when the rotational speed of the rotor is larger than a predetermined rotation A control circuit for a sensorless brushless motor, wherein current is supplied to the motor winding at.
The current supply means allows a minute current to flow through the plurality of motor windings in a predetermined order for a predetermined time when the mode is switched from the first mode to the second mode. A control circuit for a sensorless brushless motor according to claim 8.
DC cut-off means capable of switching between a first cut-off frequency and a second cut-off frequency greater than the first cut-off frequency, and removing a direct-current component superimposed on the magnetic flux signal;
Switching means for switching between the first cutoff frequency and the second cutoff frequency of the DC cutoff means;
The sensorless brushless motor control circuit according to claim 8 or 9, further comprising:
The switching means sets the cut-off frequency of the DC cut-off means as the first cut-off frequency for a predetermined time after the rotor starts, and switches to the second cut-off frequency when a predetermined time elapses. The sensorless brushless motor control circuit according to claim 10.
A rotor with magnetic poles;
A motor unit comprising a plurality of motor windings for rotating the rotor;
Current supply means for supplying current to the plurality of motor windings;
Magnetic flux acquisition means for acquiring an interlinkage magnetic flux of at least one of the motor windings by the magnetic pole;
Magnetic pole position acquisition means for acquiring the position of the magnetic pole from the change in the interlinkage magnetic flux acquired by the magnetic flux acquisition means;
Comprising a control unit consisting of
The magnetic flux acquisition means includes a voltage processing calculation means for performing a voltage processing calculation for subtracting a voltage based on a voltage drop due to at least a resistance from a line voltage of the motor winding, and the interlinkage when performing the voltage processing calculation. and a DC blocking unit for removing a DC component superimposed from a signal for calculating a magnetic flux,
The sensorless brushless motor apparatus, wherein the current supply means switches the current of the motor winding based on the magnetic pole position acquired by the magnetic pole position acquisition means.
A rotor with magnetic poles;
A plurality of motor windings for rotating the rotor;
Among the plurality of motor windings, a voltage difference between two predetermined phases so that the phase and magnitude of the voltage drop due to the inductance of the motor winding are equal to each other , subtraction of voltage based on the voltage drop due to resistance, and integration and arithmetic processing consisting of a magnetic flux signal acquiring means for acquiring a magnetic flux signal,
When performing the arithmetic processing, a DC blocking unit that removes a DC component superimposed on the voltage difference signal; and
From the arithmetic processing magnetic flux signals on the basis of a signal from the removed DC component superimposed by the DC blocking portion, the first driving for obtaining the drive timing of the part of the drive voltage vector of the output possible driving voltage vector Timing acquisition means;
First drive voltage vector output means for outputting the partial drive voltage vector in synchronization with the drive timing acquired by the first drive timing acquisition means;
Second drive timing acquisition means for acquiring the drive timing of the output drive voltage vector by multiplying the timing obtained from the magnetic flux signal acquired by the magnetic flux signal acquisition means;
Second drive voltage vector output means for outputting the output drive voltage vector in synchronization with the drive timing acquired by the second drive timing acquisition means;
Selecting means for selecting the first drive voltage vector output means and the second drive voltage vector output means;
A sensorless brushless motor device comprising:
An exterior body in which an air inlet is formed at one end and an air outlet is formed at the other end;
A rotor rotatably supported by a magnetic bearing or a mechanical bearing in the exterior body;
A motor for rotating the rotor;
A stator disposed in the exterior body;
Comprising
A vacuum pump device comprising the sensorless brushless motor device according to claim 12 or 13.

Images