JPH06121583A - Driver for sensorless brushless motor - Google Patents

Abstract

PURPOSE:To obtain a brushless motor drive which can perform sensorless driv ing by providing a control means for comparing an induced voltage with the neutral point voltage of a motor and eliminating phase difference between the rotational position of rotor and exciting timing. CONSTITUTION:30 deg. period of induced voltage, included in a terminal voltage waveform Ew(a) of a coil corresponding with the rotational speed of a motor, is measured by means of a timer and interruption is generated. The period thus measured is then set as an exciting time which corresponds with 360 deg. induced voltage divided by 12. The 30 deg. period measured by the timer is then outputted with a next cross point 2 as a trigger point wherein timings for take-in of period, operation, and commutation are set with reference to the trigger point at all times. Furthermore, levels are compared between the terminal voltage waveform and the neutral potential during nonconduction interval using one of three-phase coils thus correcting the 30 deg. exciting period using four types of combination. This constitution provides a stabilized operation of rotor equivalent to that of a brush motor equipped with a rotational position sensor.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a drive device for a brushless motor, which is used in home electric appliances and has no position detecting sensor.

[0002]

2. Description of the Related Art In the field of home electric appliances and the like, brush life motors have come to be used along with the long life and high reliability of products. As for the driving method, many sensorless driving methods have been proposed.

A method of sensorless driving will be described below. FIG. 9 shows a schematic block diagram of a drive circuit of a conventional sensorless brushless motor. In FIG. 9, 21 is a switching circuit, 22 is a drive pulse generation logic, and 23
Is a delay circuit, 24 and 25 are comparators, and 26 is a microprocessor. La and Lb are motor coils.

In FIG. 10, (a) is an induced voltage, and (b) is
Is a pulse signal obtained from the comparator, and (c) is a delay clock. The motor used in the figure is a two-phase full-wave drive.

When the motor is rotating, each coil La,
As shown in the waveform diagram of FIG.
The induced voltages Ea and Eb having a sinusoidal phase difference are generated.
The induced voltages Ea and Eb are 2 at both ends of the coils La and Lb.
The waveform is shaped at the neutral voltage of the motor by the comparators 24 and 25 to which the inputs are connected. Therefore,
As shown in FIG. 10B, pulse signals S1 and S2 having the same period and phase as the induced voltages Ea and Eb and a phase difference of 90 degrees are obtained from the comparators 24 and 25, respectively.

The pulse signals S1 and S2 are supplied to the delay circuit 23, and as shown in FIG. 10 (c), the rising edges of S1 and S2 form a delayed clock DCK whose rising is delayed by the time T. The time T corresponds to an electrical angle of 45 degrees, and thus a 90 degree width electrical angle can be set with the front edge at a position of 45 degrees from the magnetic pole boundary that is the reference position of the rotor magnet without a rotational position sensor. Is sensorless driven.

[0007]

However, in the above configuration, since energization switching of the motor is performed with reference to the delay pulse DCK, an additional delay circuit must be provided,
If the delay pulse includes a time error, commutation cannot be performed at an appropriate timing.

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned problems of the prior art and to provide a drive device of a brushless motor capable of sensorless driving which can perform commutation at an appropriate timing without providing a delay pulse.

[0009]

In order to achieve this object, a drive device for a sensorless brushless motor according to the present invention comprises:
Starting means from the energization start state, means for measuring a cycle of the induced voltage of at least one exciting coil at an electrical angle of 30 degrees, means for switching energization of each coil based on the measured cycle, and an induced voltage Controls to eliminate the phase difference between the rotor rotation position and the excitation timing by comparing the induced voltage and the neutral point voltage of the motor with the means for switching the energization of each coil based on the position where the neutral point voltage of the motor crosses. And means for doing so.

[0010]

With the above-described structure, the present invention can generate an excitation time according to the rotation speed of the motor without using the rotation position sensor of the rotor, and can switch the energization optimally.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a drive device for a sensorless brushless motor according to the present invention. In FIG. 1, reference numeral 1 denotes a microprocessor, which determines the excitation switching timing of each coil, outputs the excitation switching signals P1 to P6 of each coil, and a cycle T (180) corresponding to an electrical angle of 30 degrees of the induced voltage of the excitation coil. Take in. Reference numeral 2 is a driver for switching the excitation of each coil, and 3 is a brushless motor in which three phases are connected to perform three-phase full-wave drive. Reference numeral 4 is a comparator for comparing the induced voltage (for example, Ew) of a specific one of the three exciting coils of the motor with the neutral point voltage of the motor, and 5 is a level conversion circuit for converting the voltage level of the output of the comparator 4. , 6 is a gate circuit for combining the output signal of the level conversion circuit 5 and the output signal of the microprocessor 1,
Reference numeral 7 is a neutral point voltage of the motor or an equivalent reference voltage.

The operation of the embodiment of the present invention will be described in more detail. First, energization of each of the three phases is forcibly started from the stopped state of the motor. This energization method is called a so-called full-wave drive method for a three-phase motor, and U → V, W → V, W → U, V → U, V → W, for each phase of U, V, W. U → W
And energize sequentially. At this time, the rotor rotates synchronously in a so-called open loop in response to this excitation.

At the time of starting, the excitation time of each excitation pattern is so long that the motor is started, and the current is energized, and the excitation time is gradually increased to increase the rotation speed of the rotor, so-called slow-up drive is performed. When the rotation speed of the rotor reaches a sufficient speed, the induced voltage reaches a level that can be processed. This state is shown in FIG.

Here, for example, focusing on the W phase, it can be seen that the cycle of the electrical angle of 30 degrees can be measured by comparing with the neutral point potential. The 30-degree cycle T (30) is measured by a microprocessor. At the time of this measurement, a voltage crossing the transient neutral point potential is generated due to switching of the exciting coil other than the original 30-degree cycle T (30). The software of the microprocessor is controlled to observe only the falling edge. Alternatively, the mask pulse is generated from the microprocessor and is removed to measure the period T (30) corresponding to the induced voltage of 30 degrees.

FIG. 2 is an example of an internal configuration diagram of the microprocessor 1 which can be used in the present invention. Output P7 of level conversion circuit and cross position detection signal P9 of induced voltage
Via the input circuit 12 to 8 central processing units (hereinafter CPU
Abbreviated). The CPU 8 is connected to the memories of 10 timers 1, 11 timers 2 and 9, respectively. Reference numeral 13 denotes an output circuit, and the output of the CPU supplies signals P1 to P6 and P8 to the driver 2 and the gate circuit 6, respectively.

FIG. 3 is a main part waveform diagram of each part. FIG. 3A shows a terminal voltage waveform of one exciting coil, which is Ew. This waveform includes an induced voltage component that is rotation information of the rotor, and is like a dotted line showing only this induced voltage component. This waveform is input to the comparator 4 of FIG. 1, compared with the neutral point voltage of the motor, and the result is input to the level conversion circuit 5. The output is shown in FIG. This waveform (b) includes the same period and phase as the induced voltage Ew.

A 30 ° cycle T (30) is from the trailing edge of the edge indicated by the arrow in FIG. 3B to the next trailing edge.

Here, using the timer 1, the induced voltage 3
Measure the 0 degree cycle. For example, as shown in FIG. 3C, a pulse signal is output in order to remove a transient voltage due to switching of the exciting coil, and a waveform that is ORed with (b) waveform and (d) waveform is used. , The period from the rising edge to the falling edge of the arrow in (d), that is, 3
The cycle T (30) corresponding to 0 degree is measured, and an interrupt is generated at the falling edge of the edge.

Alternatively, the software is controlled at the timing shown in (e), and only the permission timing is observed (b) only the falling edge of the waveform is measured, the period from the falling edge to the falling edge is measured, and the edge is The 30-degree cycle T (30) can be measured by generating an interrupt at the falling edge of.

This is a value obtained by dividing the induced voltage of 360 degrees into 12 parts. Therefore, in the case of three-phase full-wave driving, the excitation pattern is 1
There are two patterns.

Thereafter, the first step of the three-phase excitation pattern based on the measurement result is output at the timing when the U-phase power generation waveform crosses the neutral point potential from plus to minus. Thereafter, the three-phase excitation pattern is sequentially output at each of the calculated times.

This is because, for example, the microprocessor shown in FIG.
In the case of the configuration shown in (1), the 30-degree cycle T (30) measured using the timer 1 is measured by the timer 2 and an interrupt is generated at each time to sequentially output 12 patterns, which are three-phase excitation patterns.

Alternatively, even if the microprocessor has a configuration different from that of FIG. 2, the effect is the same if the time corresponding to 30 degrees can be measured.

The series of flows will be described with reference to FIG. 3. The coil terminal voltage waveform E corresponding to the rotation speed of the motor will be described.
The 30 degree cycle T (30) of the induced voltage included in w (a) is measured by the timer 1 in FIG. 2 and the interrupt 1 is generated. This means the fall of the edge in (b) or the fall of the edge in (d), which is the same as the position where the terminal voltage waveform crosses the neutral point voltage from plus to minus.

Next, the measured period is set as the excitation time.
This is a value obtained by dividing the induced voltage of 360 degrees by 12. Then, using the next cross point (▼ 2) as the trigger point,
The 30-degree cycle T (30) previously obtained by the measurement according to the excitation patterns P1 to P6 is measured by the timer 2 in FIG. 2 and output. That is, the induced voltage is measured at 30 ° time, and commutation is performed at 30 ° time in accordance with the 12 excitation patterns shown at P1 to P6 with the position where the induced voltage crosses as the trigger point (▼). The timing of cycle acquisition, calculation, and commutation is always based on the trigger point (▼). At the trigger point, 30
Although the measurement of the frequency cycle and the output of the excitation pattern are processed at the same time, the priority of the interrupt is that the interrupt by the measurement of the cycle of the induced voltage has a higher priority.

Further, using one of the three phases of the coil, the level of the terminal voltage waveform and the neutral point potential are compared in the non-energized period of the U phase, for example, and the four combinations of high and low are as follows. Then, the 30-degree excitation period is corrected.

FIG. 4 is a waveform diagram of essential parts showing the concept of correcting the 30-degree excitation period. In FIG. 4, (a) is a waveform of the terminal voltage of the exciting coil, in which the normal state, the phase lead, and the phase lag are overlapped. Note that only the induced voltage waveform is shown for phase delay and phase advance. (B) is a signal P7 obtained by level-converting the output of the comparator 4 during normal operation, and (c) is a signal P7 obtained by level-converting the output of the comparator 4 during phase advance.
Is a signal P7 obtained by level-converting the output of the comparator 4 when the phase is delayed.

For example, a combination of P7 at the time of the first excitation pattern output and P7 at the time of the sixth excitation pattern output, that is, a combination of high and low levels, that is, the coil non-energization period, the 30-degree excitation cycle is used. Is corrected, that is, the lead or lag of the phase is determined.

When the phase advances, P7 at the time of the sixth output of the excitation pattern becomes high level with respect to the normal low level. On the contrary, when the phase is delayed, P when the excitation pattern is first output
7 becomes high level with respect to the normal low level, and phase lead and lag can be detected. If it is determined that the phase is advanced, the excitation time of 30 degrees obtained by the measurement is delayed by a constant value or a value determined by calculation, and if it is determined that the phase is delayed, 30
The excitation time of degree is advanced by a fixed value or a value determined by calculation to correct the 30-degree excitation cycle, and control is performed so as to secure the optimum commutation timing.

Further, a method of controlling the phase of the rotation of the rotor and the energization timing by utilizing the 30 degree cycle time of the induced voltage itself will be described below.

In the state where there is no phase shift, the waveform is as shown in FIG. 5A, and the 30 degree period measured here is 3 for excitation.
The time coincides with the 0 degree cycle T. This is judged by software. On the other hand, when the phase advances or lags, the measured 30-degree cycle becomes shorter or longer than the exciting 30-degree cycle T, and it can be seen that a phase shift occurs.

When the phases are deviated as shown in (b), the 30 ° period cannot be observed in the waveform comparing the induced voltage and the neutral point voltage shown in (2), and the phase is immediately deviated. Recognize. At this time, if the level of the induced voltage waveform is determined with respect to the neutral point potential, it can be determined that this deviation is due to the advance of the rotor. On the contrary, when the phase of the rotor is delayed, it can be determined in the same manner.

As described above, since the 30-degree cycle also includes the phase information, the speed and phase can be controlled by the closed loop control by utilizing the 30-degree cycle.

If the motor is rotating at a constant speed, the 30-degree cycle T (30) of the induced voltage is constant. That is, the cycle T1 measured last time and the cycle T newly measured this time
2 should be equal. Therefore, the ratio ΔT between the periods T1 and T2 is calculated. At constant speed rotation, ΔT = 1. When the load changes, ΔT ≠ 1. If the absolute value of ΔT is less than a certain value, the cycle measured this time is adopted as the excitation time of 30 degrees as the data indicating the rotation state of the rotor, or a constant ratio to the 30 degree data measured last time,
Alternatively, ΔT is multiplied by a constant coefficient and adopted as the excitation time of 30 degrees. On the contrary, if the absolute value of ΔT is more than a certain value, it is judged that the cycle measured this time is different from the rotor rotation state including noise, and the data measured last time is adopted as the excitation time of 30 degrees. To do. For example,

[0036]

[Equation 1]

Then, the period measured this time is adopted as the excitation time of 30 degrees,

[0038]

[Equation 2]

If so, the previously measured cycle is adopted as the excitation time of 30 degrees. By obtaining ΔT, it is possible to prevent the time different from the rotation information of the rotor from being determined as the 30-degree cycle time.

The following processing is performed as a fail safe. Due to the characteristics of the motor, when the measured 30-degree cycle does not rotate, a cycle of restarting starting from the open loop drive or shutting off the motor power supply is performed.

Next, the masking method for removing the transient voltage generated at the time of switching the coil excitation and measuring the 30-degree cycle will be described in detail. It has already been explained that the same effect can be obtained by processing this with software or hardware.

First, the case of software processing will be described. The 11th time is output so as to observe only the trailing edge of the edge in FIG. 3B, the waveform is observed (b), and the observation is stopped after outputting the first time. Thus, the period from the trailing edge to the trailing edge of the waveform (b), that is, 3
The 0 degree period T (30) can be measured.

Next, a case of processing by hardware will be described. Twelve excitation patterns are output every 30 ° cycle measured. It is known that the transient voltage is generated immediately after the 12th output, so
The mask signal is output immediately after the twelfth output. By outputting the mask signal immediately after outputting the twelfth time, it is possible to measure the 30-degree cycle that should be originally measured with the minimum dead time. Further, the mask signal width can be made variable depending on the rotation speed of the motor and the motor voltage or motor current value. The rotation speed is proportional to the 30-degree cycle, and the current value can be estimated by the voltage and the rotation speed.
Then, this mask signal is obtained by synthesizing the waveforms of FIG. 3C and FIG. 3B to obtain FIG. 3D, and the period from the rising edge to the falling edge of FIG. This pulse width is measured.

The method of removing the transient voltage generated when the coil is switched and measuring the 30-degree cycle accurately is PWM.
It is also effective when performing speed control such as control.

FIG. 6 shows a conceptual diagram for speed control. In FIG. 6, (a) shows the terminal voltage waveform, which is the case where PWM control is performed with the clock and duty shown in (b). The waveform obtained by comparing the terminal voltage waveform and the neutral point voltage and converting the level is shown in (c). The cycle of the induced voltage of 30 degrees is the cycle indicated by the arrow in (c). Therefore, from the waveform in (c), 30
In order to detect the frequency cycle by software, in the same manner as described above, the software processing is performed so that only the falling edge corresponding to 30 degrees is observed, and the 30 degree cycle is measured.

In the case of processing by hardware, the mask signal is output immediately after the transient voltage due to the switching in the three-phase logic is generated and immediately before the transient voltage due to the switching by PWM is generated in the same manner as described above. The mask signal is output, the waveform (d) of FIG. 6 is obtained by waveform synthesis, and the period corresponding to 30 degrees is measured.

In this way, even when speed control such as PWM is performed, the induced voltage 30 ° cycle can be measured by masking by soft or hard processing.
After that, the motor is controlled in the same procedure as when the speed is not controlled.

When controlling the speed of the motor, generally, the voltage is controlled by PWM or the like from several hundreds of revolutions to several thousands of revolutions. However, when the speed is controlled by the above-mentioned mask method, the cycle is 30 degrees at a very low revolution speed. There are some cases where you can't get exactly. Further, even if it is accurately obtained, when it is attempted to operate the motor at a rotation speed that is determined by the voltage around the so-called minimum operating voltage of the motor and the load, the rotation of the motor becomes unstable and is likely to get out of step.

Therefore, when the load is light and the rotation speed is extremely low, synchronous rotation is performed by a so-called open loop forcibly energizing, which is the energizing method for starting described above. In other words, since the microprocessor knows the number of rotations of the motor, it rotates synchronously in an open loop below a certain number of rotations, measures a 30 degree cycle above a certain number of rotations, and rotates at an excitation cycle according to that cycle. It is also possible to use differently.

FIG. 7 shows the range of speed control. The range of the arrow, that is, the range from the maximum rated voltage to the minimum starting voltage is speed-controlled by PWM or the like. However, even when the speed is less than the minimum starting voltage, that is, the speed is controlled by the shaded area in FIG. If it is set higher and is rotated synchronously in the open loop described above, it becomes possible to rotate at an extremely low speed without step out.

As described above, when the rotation speed is lower than a certain value, the motor is synchronously rotated in an open loop, and the voltage applied to the motor at that time is set higher than the minimum operating voltage. By measuring a 30 degree cycle above a certain number of rotations and using them so that they rotate at an excitation cycle according to that cycle, the speed can be controlled from a rotation speed lower than that at the minimum starting voltage, and stable speed control can be achieved over a very wide range. It becomes possible to do.

FIG. 8 shows a flowchart of an embodiment in which the microprocessor having the structure shown in FIG. 2 is used.

Although the embodiment in which the present invention is applied to the three-phase full-wave drive type brushless motor has been described above, the present invention can be applied to a half-wave or full-wave drive type sensorless brushless motor of two or more phases. It is possible.

Further, an example of measuring the induced voltage of one exciting coil when measuring the period of the induced voltage of 30 degrees has been shown, but the same effect can be obtained by using two or three exciting coils. To be In this case, since the controllability of the closed loop becomes stronger as the number of exciting coils is increased, it is possible to use the closed coil properly depending on the application.

The same effect can be obtained by measuring the period from the rising edge of the induced voltage to the rising edge.

Further, an embodiment has been shown in which the excitation is switched based on the position where the W-phase power generation waveform crosses the neutral point potential from plus to minus with respect to the reference position of the commutation timing of the motor. If it is a motor, U,
It goes without saying that the same effect can be obtained by switching the excitation with reference to any one of the positions where the generated voltage of either the V coil or the W coil crosses the neutral point potential.
For example, when the above-mentioned three-phase 30-degree cycle is measured and used for control, it is also possible to control such that the calculation result is immediately output from the next midpoint potential cross.

In the above description, an example in which the excitation pattern is 12 patterns corresponding to a 30 degree cycle has been explained.
6 patterns corresponding to 0 degree cycle or excitation 360
It goes without saying that the beginning and the end of the degree can be treated as a 30 degree cycle, a 60 degree cycle, or a mixed type thereof.

Further, although the start-up by the slow-up has been described as the control at the time of start-up, when the so-called start-up torque of the motor is sufficiently large with respect to the load, the slow-up is omitted and the induced voltage immediately reaches a sufficient level for control. Needless to say, it can be started from the number of revolutions generated in.

[0059]

As described above, according to the present invention, the electrical angle 30 degree period T (30) of the induced voltage is measured, and the cross point of the induced voltage is used as a trigger point to perform commutation in 30 degree time according to the excitation pattern. When using the measured 30 degree cycle T (30), the ratio of the cycle measured last time and the cycle newly measured this time is taken, and if it is a certain value or more, the previous data is adopted to calculate the excitation time, In addition, by determining the signal level obtained by converting the output of the comparator at a specific timing and performing phase lock control, stable operation equivalent to that of a brushless motor with a rotor rotation position sensor can be obtained.

[Brief description of drawings]

FIG. 1 is a block diagram of a drive circuit according to an embodiment of the present invention.

FIG. 2 is an internal configuration diagram of a microcomputer according to an embodiment of the present invention.

FIG. 3 is a waveform diagram of essential parts in an embodiment of the present invention.

FIG. 4 is a conceptual diagram of phase lock control in one embodiment of the present invention.

FIG. 5 is an operation waveform diagram of phase control in one embodiment of the present invention.

FIG. 6 is a conceptual diagram of speed control in one embodiment of the present invention.

FIG. 7 is a diagram illustrating a range of speed control according to an embodiment of the present invention.

FIG. 8 is a flowchart in an embodiment of the present invention.

FIG. 9 is a schematic block diagram of a drive circuit of a conventional sensorless brushless motor.

FIG. 10 is a main part waveform diagram of a conventional sensorless brushless motor.

[Explanation of symbols]

 1 Microprocessor 2 Driver 3 3-phase brushless motor 4 Comparator 5 Level conversion circuit 6 Gate circuit 7 Motor neutral voltage 8 Central processing unit (CPU) 9 Memory 10 Timer 1 11 Timer 2 12 Input circuit 13 Output circuit 21 Switching circuit 21 22 drive pulse generation logic 23 delay circuit 24, 25 comparator 26 microprocessor

Claims (9)

[Claims] 1. A means for measuring the period of an electrical angle of 30 degrees of an induced voltage included in a terminal voltage generated at a coil terminal by using a terminal of an exciting coil, and switching of energization of the coil based on the measured period. A sensorless brushless motor drive device including a means for performing the operation. 2. The coil energization switching is performed based on the position where the induced voltage of the exciting coil is compared with the neutral point voltage of the motor and the position where the plus crosses the minus or the minus crosses the plus. The drive device for the sensorless brushless motor according to claim 1. 3. A mask for detecting only a specific timing when detecting the position where the induced voltage of the exciting coil crosses the neutral point voltage of the motor.
A drive device for the sensorless brushless motor described. 4. When using the cycle of the electrical angle of 30 degrees of the induced voltage of the exciting coil, take the ratio or difference between the cycle previously measured and the cycle newly measured, and if the ratio or difference is not less than a specific ratio or difference, 3. The sensorless brushless motor driving device according to claim 2, wherein the coil excitation time is determined by using the measured period. 5. A level determination is performed at a specific timing by using a signal level obtained by comparing an induced voltage of an exciting coil with a neutral point voltage of a motor, and a coil energization switching timing and
The drive device for the sensorless brushless motor according to claim 2, wherein control is performed so as to eliminate a phase difference from the rotor position. 6. The sensorless brushless motor driving device according to claim 2, wherein a mask pulse signal is output in order to suppress a counter electromotive voltage generated when switching the energization of the coil. 7. The drive of a sensorless brushless motor according to claim 3, wherein the mask pulse signal width for suppressing the back electromotive voltage is changed in accordance with either or both of the rotation speed of the motor and the motor current. apparatus. 8. After starting in an open loop at the time of start-up and gradually increasing the rotation speed so that the induced voltage of the exciting coil can be detected,
The drive device for a sensorless brushless motor according to claim 2, wherein closed-loop control is performed. 9. A drive device for a sensorless brushless motor according to claim 1, wherein when performing speed control, open-loop drive and closed-loop drive are selectively used according to the rotational speed.