JP2001327185A - Method and apparatus for detecting initial position of dc brushless motor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and an apparatus capable of easily detecting a position of a rotor at a starting time without using a sensor. SOLUTION: An AC voltage is applied to a U phase of a DC brushless motor 1. Thus, an exciting current Iu flowing in this case and induced voltages Vv, Vw generated in the other two phases are measured. Since the Iu, Vv and Vw are changed due to bids magnetization phenomenon of a permanent magnet by the position of the rotor, the initial position of the rotor can be detected based on these three pieces of information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and a device for detecting an initial position of a rotor in a DC brushless motor, and more particularly to a method and a device capable of detecting a rotor position without using a sensor. is there.

[0002]

2. Description of the Related Art A DC brushless motor is a motor in which a commutator and a brush in a permanent magnet type DC motor are replaced by an electronic commutation mechanism, and generates no noise or noise because there is no mechanical contact portion. And has a long service life. In addition, they are widely used in the industry because they are more efficient than induction motors, are small, and are relatively easy to control.

[0003] Since this DC brushless motor is structurally a synchronous motor using a permanent magnet for a rotor, in order to drive the motor, it is necessary to supply a current corresponding to the position of the magnet to the coil. For this reason, means for detecting the initial position of the rotor is required. Conventionally, as the initial position detecting means, a Hall element sensitive to a magnet has been often used.

[0004]

However, although the size of the Hall element is small, its occupancy cannot be ignored in reducing the size of the motor, which is a hindrance to the miniaturization. There is also a problem that the driving characteristics of the motor are affected by the mounting accuracy of the Hall element. further,
In order to detect the rotor position with high accuracy, many Hall elements are required, the number of signal lines from the motor to the drive circuit is increased, and a large wiring area is required on the printed circuit board. In addition, there is an optical sensor using an optical sensor in addition to the Hall element.

Therefore, if a brushless DC motor can be driven without using a sensor for position detection,
Not only is it possible to further reduce the size and cost of the motor, but it is also free from problems such as deterioration of the Hall element and the like and disconnection of the signal line, thereby realizing improved reliability.

By the way, when the motor is rotating, an induced voltage is generated by the permanent magnet field of the rotor, and by using this, the position of the rotor can be detected relatively easily. For example, Japanese Patent Publication No. 59-365
Japanese Patent Application Laid-Open No. 19-1990 and Japanese Patent Application Laid-Open No. 3-239186 disclose a technique for detecting the position of a rotor by an induced voltage generated in a stator winding or an armature winding.

However, in the above-mentioned publication, the position is detected while the motor is started and the rotor is rotated. Therefore, this method is directly applied to the position detection at the time of starting when the rotor is stationary. And it is not easy to detect the position of the rotor with high accuracy.

It is an object of the present invention to provide a method and an apparatus which can easily detect the position of a rotor at the time of starting without using a sensor.

[0009]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention focuses on the phenomenon of demagnetization caused by permanent magnets.
By using the fact that when one phase is excited, the induced voltage induced in the other phase and the exciting current become non-uniform due to the magnetic polarization phenomenon, the rotor magnetic pole position is detected. is there.

Before going into the details of the present invention, the basic principle of the present invention will first be described. When the motor is stopped, a DC magnetic flux passing through the iron core is generated by the magnetic field generated by the permanent magnet of the rotor. Generally, when a DC magnetic flux passes through the iron core, the operating point of the hysteresis loop deviates from the origin of the initial magnetization characteristic curve. Therefore, if an AC magnetic field is applied in this state, the exciting current flowing through the coil becomes asymmetric, so that the hysteresis loop becomes asymmetric, that is, a so-called DC bias phenomenon occurs. When a constant AC voltage is applied in this state, the larger the amount of magnetic flux passing through the iron core, the larger the current flowing to excite the iron core, and at the same time the peak position of the current rises and falls depending on the direction of the magnetic flux. shift. Therefore, even when an AC voltage is applied to one phase of the motor, a magnetic demagnetization phenomenon appears. Further, since the amount and direction of the magnetic flux passing through the iron core differs depending on the position where the permanent magnet of the rotor is stopped, the above-described change is seen in the exciting current.

When an AC voltage is applied to one phase of the motor, an AC magnetic flux is generated, and a voltage is induced in the coils of the other two phases. At this time, a DC magnetic flux due to the magnetic field of the permanent magnet of the rotor is also generated in the iron cores of the other two phases, so that the AC magnetic flux passing through the cores of those phases is affected. For example, when a large amount of DC magnetic flux passes through the iron core, the voltage induced by the AC magnetic flux decreases, and when the DC magnetic flux passing through the iron core is small, the induced voltage increases. Therefore, depending on the position where the permanent magnet of the rotor is stopped, the amount and direction of the magnetic flux passing through the iron cores of the other two phases differ, so that the induced voltages of those phases differ.

According to the present invention, when an AC voltage is applied to a certain phase as described above, both the exciting current and the induced voltage of the other phase change depending on the position where the permanent magnet of the rotor is stopped. Based on this, the initial position of the rotor is detected.

[0013]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described.
This will be described with reference to the drawings. FIG. 1 shows a basic circuit used in the initial position detecting method of the present invention. In the figure, 1 is a DC brushless motor, 11, 12, and 13 are U-phase, V-phase, and W-phase coils, respectively, 2 is an AC power supply, and 3 is an ammeter. Iu is the U-phase excitation current, and Vv is V
The voltage induced in the phase coil 12 and Vw represent the voltage induced in the W phase coil 13. An AC voltage is applied to the U phase of the DC brushless motor 1 by the AC power supply 2, and an exciting current Iu flowing at that time and induced voltages Vv and Vw generated in the other two phases are measured. The initial position is detected based on the information. The details will be described later.

In FIG. 1, as a DC brushless motor,
The following specifications are used. Number of poles: 8 poles Rated output: 600 W Rated voltage: 48 V Rated rotation speed: 800 rpm Rated torque: 7.5 Nm Armature resistance: 22 mΩ Inductance: 73.25 μH Permanent magnet material: neodymium magnet The voltage of the AC power supply 2 is 2. 5V, frequency is 60Hz
And

FIG. 2 is a diagram showing the arrangement of the coils and the rotor of the DC brushless motor 1. As shown in FIG. Reference numerals 11, 12, and 13 denote the above-described U-phase, V-phase, and W-phase coils, 4 denotes a stator core,
5 is a rotor. The rotor 5 is composed of a permanent magnet in which N poles and S poles are alternately arranged. In the following description,
The state in FIG. 2 is a reference position where the mechanical angle is 0 °, and a counterclockwise direction from this position is a rotation direction.

In the circuit shown in FIG. 1, when an AC voltage is applied to the U-phase, how the Iu and Vv, Vw change while the rotor 5 is moved by one pole pair (90 ° in mechanical angle). Was measured at every mechanical angle of 1 °. When the rotor 5 is stopped, the mechanical angles are “0 °”, “22.5 °”,
The actual measurement results and the magnetic field analysis results at “45 °” and “67.5 °” are shown in FIGS. In each figure, (a) is a schematic diagram of a positional relationship between a stator and a rotor,
(B) is an actual measured excitation current waveform and induced voltage waveform,
(C) shows the result of the magnetic field analysis by the finite element method.
The results of the magnetic field analysis are based on a simple model that can be considered equivalent to the structure of the DC brushless motor 1 used.
In the state where an AC voltage is applied to the U phase, analysis is performed by the finite element method.

As can be seen from FIGS. 3 to 6, the exciting current Iu and the induced voltages Vv and Vw change depending on the position of the rotor 5 due to the magnetic polarization phenomenon. Here, FIG. 4C and FIG.
In (c), most of the magnetic flux generated by the magnetic field of the permanent magnet passes through the U-phase iron core. Therefore, at these positions, the operating point of the hysteresis loop is greatly shifted from the origin of the initial magnetization characteristic curve under the influence of the magnetic flux of the permanent magnet. As a result, when an AC voltage is applied, a large excitation current Iu flows as shown in FIGS. 4B and 6B.

FIG. 4B shows that the mechanical angle is “22.5
At the position “°”, the peak position of the waveform of the exciting current Iu is shifted in the positive direction due to the influence of the magnetic flux generated from the N pole. That is, it can be seen that the non-uniformity of the exciting current due to the generally-used magnetization phenomenon also appears inside the motor. On the other hand, the mechanical angle “67.5
6B, the peak position of the waveform of the exciting current Iu is shifted in the negative direction due to the influence of the magnetic flux flowing into the S pole as shown in FIG. 6B. The above matters are important in accurately detecting the position of the rotor 5 as described later. Further, at mechanical angles “0 °” and “45 °” through which the magnetic flux by the permanent magnet hardly passes, the excitation current Iu is hardly affected by the magnetic flux.
5 (b) and a symmetrical waveform as shown in FIG. 5 (b).

As described above, the exciting current Iu at each position of the rotor 5 was measured, and the rotor 5
°) Plotting the change of the maximum value and the minimum value of the exciting current Iu during the movement, the result as shown in FIG. 7 is obtained.

Next, the stop position of the rotor 5 is changed to the mechanical angle "1".
The measured results and the magnetic field analysis results at 5 °, 30 °, 60 °, and 75 ° are shown in FIGS.
Shown in In each of the figures, as in FIGS. 3 to 6, (a) is a schematic diagram of the positional relationship between the stator and the rotor, (b) is an exciting current waveform and an induced voltage waveform by actual measurement, and (c) is a finite element method. 3 shows the results of a magnetic field analysis by the method shown in FIG.

Here, in FIG. 8 (b), the induction of the W-phase through which the magnetic flux of the permanent magnet passes more than the induction voltage Vv of the V-phase through which the magnetic flux of the permanent magnet does not pass much. The voltage Vw is very small. 9 to 11, a similar phenomenon is observed, and the induced voltage of the phase through which the magnetic flux of the permanent magnet passes is smaller than the induced voltage of the phase through which the magnetic flux does not pass much. This is considered to be due to the fact that the AC magnetic flux generated from the stator winding is unlikely to flow into the iron core, which has a large amount of magnetic flux due to the permanent magnet that prevents the magnetic flux.

It can also be seen that the manner in which the voltage is induced differs depending on whether the AC magnetic flux generated from the stator winding is in the same direction as the magnetic flux of the permanent magnet or in the opposite direction. FIG. 1 shows how the maximum values of the induced voltages Vv and Vw change while the rotor 5 moves by one counter electrode (90 °).
It looks like 2.

From the above, it has been found that the exciting current and the induced voltage change depending on the position of the rotor 5, and a method of detecting the initial position of the rotor 5 based on this phenomenon will be described below. I do.

First, attention is paid to a change in the induced voltage.
FIG. 12 shows that the maximum values of the voltages Vv and Vw induced in the V and W phases change by two periods while the rotor 5 moves by one counter electrode. Therefore, for example, the mechanical angle “15”
° ”and a mechanical angle“ 60 ° ”that is just one pole away from it.
, The maximum values of Vv and Vw are almost the same value (the maximum value of Vv is about 1.6 V at 15 ° and 60 °, and the maximum value of Vw is about 0.1 V at 15 ° and 60 °). 5V). Therefore, only the information on the peak values of the induced voltages Vv and Vw can detect the angle of the magnetic pole of the rotor 5, but cannot determine the polarity.

This problem can be overcome by using the information of the exciting current Iu. FIG. 8B and FIG.
Paying attention to the waveform of the induced voltage Vw in (b), it can be seen that the position where the value of Vw is maximum is different between the two.
That is, the position where the induced voltage Vw is maximum is as shown in FIG.
In the case of the mechanical angle “15 °” in (b), the exciting current Iu is negative, and in the case of the mechanical angle “60 °” in FIG. 10B, the exciting current Iu is positive. Similarly, the position of the mechanical angle “30 °” and the position of the mechanical angle “75 °” which is shifted by one pole from the mechanical angle are shown in FIGS.
As can be seen from FIG. 1B, the position where the value of the induced voltage is maximum is different. Therefore, the polarity of the magnet can be determined by using such a relationship between the peak position of the induced voltage waveform and the exciting current. Therefore, the case where the position of the maximum value is as shown in Table 1 is considered.

[Table 1]

While the rotor 5 moves one counter electrode, the excitation current Iu and the induced voltages Vv, Vw measured at each position are measured.
Is shown in FIG.
The vertical axis of FIG. 13 represents the case classification numbers in Table 1 above. In FIG. 13, mechanical angles “20 °” to “30 °”
And the mechanical angle "6
5 ° ”to a certain position in a range excluding“ 75 ° ”;
It can be seen that the case number at the position shifted by one pole from the position is different. Therefore, even if the maximum value of the induced voltage becomes the same at the mechanical angles of “15 °” and “60 °” as described above, the position can be distinguished by the case number in Table 1 ( In the case of 15 °, the classification number is 1, and in the case of 60 °, the classification number is -1). In this way, the cases shown in Table 1 are performed, and the initial position of the rotor can be detected from the combination of the change in the induced voltage and the change in the exciting current.

In FIG. 13, the mechanical angle "20"
“°” to “30 °” and the mechanical angles “65 °” to “75 °” that are shifted by exactly one pole with respect to the position, have the same case number. Therefore, an accurate position cannot be detected only in these areas by using the reference in Table 1. Therefore, the present invention overcomes this problem by utilizing the above-described shift of the peak position of the exciting current Iu.

That is, the mechanical angle “20 °” to “30 °”
In the range from “65 °” to “75 °”, the N and S poles of the magnet of the rotor 5 are located near the center of the U-phase coil, and the magnetic flux generated by the magnet passes through the U-phase core. (See FIG. 4 (c) and FIG. 6 (c)). As a result, in these ranges, the peak value of the exciting current Iu shifts remarkably. That is, as shown in FIG. 4B, the mechanical angle “2”
At the position of “2.5 °”, the peak position of the exciting current Iu shifts in the positive direction due to the influence of the magnetic flux generated from the N pole of the magnet. On the other hand, as shown in FIG.
At the position of “5 °”, the peak position of the exciting current Iu shifts in the negative direction due to the influence of the magnetic flux flowing into the S pole of the magnet. Therefore, by determining in which direction the peak position of the exciting current Iu is shifted in the positive or negative direction,
The position of “22.5 °” and the position of “67.5 °” can be distinguished to determine the polarity of the magnet.

With the above method, it is possible in principle to detect all rotor positions. However, the above description is for the case where an AC voltage is applied to the U phase. For more accurate initial position detection, it is necessary to apply the AC voltage to the V phase and the W phase and perform the same measurement. But,
This will be described later.

Incidentally, a series of procedures for initial position detection according to the above-described principle can be realized by software processing using a microcomputer, for example. For this reason, it is necessary to first make a database of the relationship between the rotor position (actually measured angle), the excitation current and the induced voltage. Therefore, in order to make the calculation simple and accurate, the following four types of data are extracted from the information on the excitation current and the induced voltage. Data1: Maximum value of Vv−Maximum value of Vw Data2: Number obtained by dividing the maximum value of Vv and Vw Data3: Maximum value of Iu + Minimum value of Iu Data4: Maximum value of Iu + Absolute value of minimum value of Iu

FIG. 14 shows how each of the above data changes during the one-pole movement of the magnet of the rotor 5. Each data in FIG. 14 is stored in the memory as a database. Here, since Data1, Data3, and Data4 change periodically, in order to simplify the configuration of the database, it is preferable to convert these data into functions using Fourier series expansion.

FIG. 15 shows a procedure for detecting the initial position of the rotor 5 using the database as described above. This procedure is referred to as method1, which will be described below.
First, an AC voltage is applied to the U phase, and the exciting current I
The values of u and the induced voltages Vv and Vw are read into a computer (step S1). Then, the read Iu, V
The following four types of data are extracted from the values of v and Vw (step S2). Data 1: Maximum value of Vv−Maximum value of Vw Data 2: Number that classifies the maximum value of Vv and Vw Data 3: Maximum value of Iu + Minimum value of Iu Data 4: Maximum value of Iu + Minimum value of Iu value

Next, the value of the extracted data 1 is compared with Data1 of the functionalized database of FIG. 14 to select a candidate angle (step S3). Since there are a plurality of angles corresponding to a certain value of data 1, a plurality of candidate angles are selected. Subsequently, the value of the data 2 at the angle selected as a candidate is compared with the value of Data2 in the database to narrow down the candidate angle (step S).
4). Here, the candidate angles are narrowed down by the case classification numbers described above. Thereafter, the values of the actually measured data 3 and data 4 are compared with the values of Data3 and Data4 in the database for the angles remaining as candidates (step S5). Here, the above-described exciting current I
The candidate angles are narrowed down by the information on the peak position shift of u. And finally, Data3, Data4
And the angle with the smallest error is selected as the initial position (step S6). By the above procedure, the initial position of the rotor 5 can be detected.

FIG. 16 is a graph showing the degree of coincidence between the detected angle (vertical axis) and the measured angle (horizontal axis) when the initial position is detected by applying an AC voltage to the U phase. FIG.
From this, it can be seen that the angle is detected with very high accuracy at most positions. However, a detection error has occurred at two points of the mechanical angles “51 °” and “83 °”. The detection angles at these positions are shifted by approximately one pole compared to the actual angles. When the detection is further repeated in the experiment, the mechanical angle “7”
°, 38 °, 52 °, and 83 °, several detection errors occurred. The angles erroneously detected at these positions were all shifted by almost one pole from the actual angles.

The cause of the above-described detection error is that the waveform of the exciting current and the waveform of the induced voltage are very similar at a specific angle. For example, the waveforms of the exciting current and the induced voltage at the mechanical angle of “7 °” are as shown in FIG. 17A, and the waveforms of the exciting current and the induced voltage at the mechanical angle of “52 °” are as shown in FIG. 17B. Become. As is clear from this, the respective waveforms of both are almost the same. FIGS. 18A and 18B show the waveforms of the excitation current and the induced voltage when the mechanical angles are “38 °” and “83 °”, respectively.
Shown in Again, the waveforms of both are very similar.

As described above, according to the procedure of Method 1, it is possible to perform almost correct detection at almost all angles. However, since there is a position where a detection error occurs as described above, the motor is only used in this procedure. Driving leaves a problem in terms of reliability. Therefore, in the present invention, when the detected angle is a position where erroneous detection is likely to occur, the excitation phase is switched and the angle is detected again. Before explaining this,
The characteristics when the W phase and the V phase are excited will be described.

FIG. 19 shows a relationship between four types of data and an angle, which is a database for detecting an initial position in the W phase using the same method as described above. W
The phases are shifted by a mechanical angle of 30 ° with respect to the U phase (see FIG. 2). Since each phase has periodicity due to the structure of the motor, FIG. 19 shows a state where the mechanical angle is shifted by 30 ° as compared with FIG.

FIG. 20 shows the degree of coincidence between the detected angle and the measured angle when an AC voltage is applied to the W phase and the initial position is detected using the database in FIG. Also in FIG. 20, it can be seen that the angle is detected with very high accuracy at most positions. However, detection errors have occurred at two points of mechanical angle “20 °” and “83 °”. Further, as a result of repeated detection, several detection errors occurred near the mechanical angles “20 °”, “38 °”, “65 °”, and “83 °”. The angles erroneously detected at these positions are almost one pole apart from the actual angles, similarly to the detection errors in the U phase. The cause of these detection errors is the same as in the case of the U phase, because the waveform of the exciting current and the waveform of the induced voltage are very similar at a specific angle.

FIG. 21 shows a relation between four types of data and an angle, which is a database for detecting the initial position in the V phase. The V phase is shifted by 60 ° in mechanical angle from the U phase (see FIG. 2). Since each phase has periodicity due to the structure of the motor, it can be seen that FIG. 21 is shifted by 60 ° in mechanical angle from FIG.

FIG. 22 shows the degree of coincidence between the detected angle and the actually measured angle when an AC voltage is applied to the V phase and the initial position is detected using the database shown in FIG. Also in FIG. 22, it can be seen that the angle is detected with very high accuracy at most positions. However, a detection error has occurred at the position of the mechanical angle “67 °”. As a result of repeated detection, the mechanical angles “7 °” and “21” were obtained.
°, 52 ° and 66 °, several detection errors occurred. The angles erroneously detected at these positions were almost one pole apart from the actual angles. The cause of the detection error is the same as in the case of the U phase.

From the above results, it can be understood that the following facts exist with respect to positions where detection errors easily occur in each phase. That is, when the initial position is detected in the U phase, detection errors are likely to occur near the mechanical angles “7 °” and “52 °”, but when these positions are detected in the W phase, a detection error occurs. Without this, it can always be detected correctly. Further, when the initial position is detected in the U phase, the mechanical angle around “38 °” and “83”
The detection error is likely to occur near "°", but when these positions are detected in the V phase, no detection error occurs and the detection can always be performed correctly.

From the above, the following three points can be said with respect to a position where a detection error easily occurs. (1) The position where a detection error easily occurs is always limited to a fixed narrow range. (2) If the detection angle is wrong, this detection angle always deviates from the actual angle by almost one pole. (3) Even in a position where a detection error is likely to occur, a correct detection can always be performed in any of the three phases.

Therefore, if the angle detected in the U-phase is included in a range in which a mistake is likely to occur when the detection is performed in the U-phase, the V-phase or W-phase can always correctly detect the angle in the range. By detecting the initial position again in the phase, the initial position can always be accurately detected.

FIG. 23 shows a flowchart of a program capable of always accurately detecting the initial position based on the above considerations. This procedure is method
2, which will be described below. First, an AC voltage is applied to the U phase (step S11), and the aforementioned meth
od1 (FIG. 15), the angle θ of the initial position
(U) is detected (step S12). Next, the detected angle θ (u) is changed from the mechanical angle “15 °” to “30”.
° ”or“ 60 ° ”to“ 75 ° ”(step S13). θ (u)
Is included in these ranges (step S13; Y
ES), since there is no risk of erroneous detection of this angle,
Let θ (u) be the angle of the final initial position (step S19).

On the other hand, if the detected angle θ (u) is not included in any of the mechanical angles “15 °” to “30 °” and “60 °” to “75 °” (step S13) ;
NO), since there is a risk of erroneous detection of this angle, θ (u) is not set to the final angle, and θ (u) is set to “0 °” to “15 °” or “45 °”. ~ "60
° ”(step S14). If θ (u) is included in these ranges (step S14; YES), an AC voltage is applied to the W phase in which there is no possibility of erroneous detection in the range to excite (step S15), and the method 1 described above is applied. Follow the steps
The angle θ (w) of the initial position is detected (Step S1)
6). Then, this angle θ (w) is used as the final initial position angle (step S19).

Also, when θ (u) is “0 °” to “15 °”,
If it is not included in any range from “45 °” to “60 °”, re-detection in the W phase may cause erroneous detection. To excite (step S17),
According to the procedure of d1, the angle θ (v) of the initial position is detected (Step S18). Then, this angle θ (v) is used as the final initial position angle (step S1).
9).

As described above, according to the detected angle,
By switching the phase to which the AC voltage is applied, the initial position can always be accurately detected. Phase switching
Although this can be performed automatically using a semiconductor switch or the like, it may be performed by manually changing the connection.
The range in which a detection error is likely to occur is actually 2 to 3 ° near the above-described four locations. In consideration of the periodicity that occurs, each section is set to a mechanical angle of 15 °.

FIG. 24 shows the degree of coincidence between the detected angle and the actually measured angle when the initial position is detected by the above method, and FIG. 25 shows the error distribution.
From these results, it can be seen that the initial position can be detected with very high accuracy with an error range within 0.5 ° at all positions. By transferring the detected angle of the initial position to the program for driving the motor, the motor can be driven without using a sensor for detecting the position.

FIG. 26 is a block diagram showing an example of the initial position detecting device according to the present invention. In the figure, 1 is the aforementioned DC brushless motor, 20 is a three-phase inverter serving as an AC power supply for supplying an AC to the coil of the DC brushless motor 1, 21 is a DC power supply switched by the three-phase inverter 20, and 22 is This is an inverter drive circuit that drives the three-phase inverter 20. Reference numeral 23 denotes a current sensor for detecting an exciting current flowing through a coil of the DC brushless motor 1, and reference numeral 24 denotes a current detection circuit for converting the current detected by the current sensor 23 into a voltage value.
3 and the current detection circuit 24 constitute a current measuring means. Reference numeral 25 denotes a voltage detection circuit that detects a voltage induced in the coil of the DC brushless motor 1, and constitutes a voltage measurement unit. Reference numeral 26 denotes a CPU to which the outputs of the current detection circuit 24 and the voltage detection circuit 25 are inputted and which controls the inverter drive circuit 22 and constitutes a rotor position detection means. Reference numeral 27 denotes a memory connected to the CPU 26, which stores the aforementioned functionalized database.

In FIG. 26, the CPU 26 detects the position of the rotor 5 in a stationary state according to the flowcharts shown in FIGS. That is, an AC voltage is applied to one phase of the DC brushless motor 1 by the three-phase inverter 20, and the exciting current at this time is measured by the current sensor 23 and the current detection circuit 24. In addition, an induced voltage generated in another phase due to the application of the AC voltage is measured by the voltage detection circuit 25. These measurement results are stored in the CPU
The CPU 26 compares these measured values with data in a database stored in the memory 27 to determine the position of the magnet of the rotor 5. Then, the inverter driving circuit 22 is controlled so that a current corresponding to the position flows through the coil of the DC brushless motor 1.

The above-described initial position detecting method and detecting device of the present invention are applied to, for example, a motor of a forklift or an automatic guided vehicle. In this case, since a sensor for position detection is not required, cost can be reduced, and the motor itself can be reduced in size and wiring can be omitted. Also, the practical effect is large in securing the degree of freedom in the layout of the vehicle body. In the case of using a position detection sensor, since the sensor itself is deteriorated,
Although it has been difficult to mount it on a forklift or the like which is often used in a bad environment, the present invention can eliminate such a problem.

[0052]

According to the present invention, the magnetic pole position of the rotor is detected by utilizing the change in the induced voltage and the exciting current due to the magnetic polarization phenomenon. And can easily be detected.

[Brief description of the drawings]

FIG. 1 is a diagram showing a basic circuit used for an initial position detection method of the present invention.

FIG. 2 is a diagram showing an arrangement of a coil and a rotor of a DC brushless motor.

FIG. 3 is a diagram illustrating changes in an excitation current and an induced voltage at a mechanical angle of 0 °.

FIG. 4 is a diagram illustrating changes in an excitation current and an induced voltage at a mechanical angle of 22.5 °.

FIG. 5 is a diagram illustrating changes in an excitation current and an induced voltage at a mechanical angle of 45 °.

FIG. 6 is a diagram illustrating changes in an excitation current and an induced voltage at a mechanical angle of 67.5 °.

FIG. 7 is a diagram showing changes in the maximum value and the minimum value of the exciting current.

FIG. 8 is a diagram illustrating changes in an excitation current and an induced voltage at a mechanical angle of 15 °.

FIG. 9 is a diagram illustrating changes in an excitation current and an induced voltage at a mechanical angle of 30 °.

FIG. 10 is a diagram illustrating changes in an excitation current and an induced voltage at a mechanical angle of 60 °.

FIG. 11 is a diagram illustrating changes in an excitation current and an induced voltage at a mechanical angle of 75 °.

FIG. 12 is a diagram showing a change in the maximum value of an induced voltage.

FIG. 13 is a diagram showing the results of classifying the excitation current and induced voltage at each position.

FIG. 14 is a diagram showing a database when detection is performed in the U phase.

FIG. 15 is a flowchart showing a procedure for detecting an initial position.

FIG. 16 is a graph showing the degree of coincidence between the detected angle and the measured angle in the U phase.

FIG. 17 is a waveform diagram of an exciting current and an induced voltage at mechanical angles of 7 ° and 52 °.

FIG. 18 is a waveform diagram of an exciting current and an induced voltage at mechanical angles of 38 ° and 83 °.

FIG. 19 is a diagram showing a database when detection is performed in the W phase.

FIG. 20 is a graph showing the degree of coincidence between the detected angle and the actually measured angle in the W phase.

FIG. 21 is a diagram showing a database when detection is performed in the V phase.

FIG. 22 is a graph showing the degree of coincidence between the detected angle and the actually measured angle in the V phase.

FIG. 23 is a flowchart for accurately detecting an initial position.

FIG. 24 is a graph showing the degree of coincidence between the detected angle and the actually measured angle.

FIG. 25 is a diagram showing an error distribution.

FIG. 26 is a block diagram illustrating an example of an initial position detection device according to the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 DC brushless motor 2 AC power supply 3 Ammeter 4 Stator core 5 Rotor 11-13 Coil 20 Three-phase inverter 23 Current sensor 24 Current detection circuit 25 Voltage detection circuit 26 CPU 27 Memory Iu Excitation current Vv, Vw Induction voltage

Claims (6)

[Claims] 1. A DC brushless motor 1 in a stationary state.
An AC voltage is applied to the phase coil, the excitation current flowing through the coil is measured by applying the voltage, and the induced voltage induced in the other phase coil is measured. And detecting an initial position of the rotor based on a combination of changes in the induced voltage. 2. The relation between each position of the rotor and the exciting current and the induced voltage is previously stored in a database, and the initial position of the rotor is detected by comparing the actually measured exciting current and the induced voltage with the database. Claim 1
4. The method for detecting an initial position of a DC brushless motor according to item 1. 3. The method according to claim 1, wherein the peak position of the waveform of the induced voltage induced in the coil of the other phase is classified into a positive section and a negative section of the exciting current, and the initial position of the rotor is determined with reference to the result. 3. The method for detecting an initial position of a DC brushless motor according to claim 1, wherein the initial position is detected. 4. A positive or negative shift of a peak value of an exciting current when a relationship between an induced voltage and an exciting current at a certain position and a position shifted by one pole from this position belongs to the same case. 4. The method for detecting an initial position of a DC brushless motor according to claim 3, wherein the initial position of the rotor is detected by determining a state. 5. When the initial position of the rotor obtained as a result of one-phase excitation is a value in a specific range in which erroneous detection is liable to occur, by exciting another phase in which erroneous detection does not occur in the range. 5. The method for detecting an initial position of a DC brushless motor according to claim 1, wherein the initial position of the rotor is detected again. 6. A DC brushless motor in a stationary state.
An AC power supply for applying an AC voltage to the phase coil; a current measuring means for measuring an exciting current flowing through the coil by applying the AC voltage; and an induction induced in the other phase coil by applying the AC voltage. Voltage measurement means for measuring a voltage, measurement results of the current measurement means and the voltage measurement means are input, and based on a combination of a change in the excitation current and a change in the induced voltage due to a stop position of the rotor, An initial position detecting device for a DC brushless motor, comprising: a rotor position detecting means for detecting an initial position.