JPH05191995A - Rotor position detector for brushless motor - Google Patents

Abstract

PURPOSE:To realize positive detection of a rotor position of a brushless motor even under low speed rotation. CONSTITUTION:Base signal of each transistor in a three-phase inverter circuit 8 for driving a motor is obtained based on the neutral point voltage VMN of a brushless motor in such a manner that a square wave output PSMN, obtained by detecting a peak position of the neutral point voltage VMN, is fed as an input signal to a three-phase ring counter circuit 7 in order to create base signals for respective transistors TUH-TWL in the three-phase inverter circuit 8.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a rotor position detecting device for a brushless motor capable of surely detecting the rotor position without a magnetic sensor even when the brushless motor rotates at a low speed.

[0002]

2. Description of the Related Art In a conventional brushless motor drive system without a magnetic sensor, an induced voltage of each phase of a three-phase coil of the motor is passed through an integrator to delay the phase and a neutral point voltage of the motor. By comparison, there is a method in which it is used as a rotor position detection signal and a current is passed through the motor three-phase coil. According to this method, since the integrator is used, the induced voltage cannot be detected when the brushless motor rotates at a low speed.

As a conventional drive system without a magnetic sensor, Japanese Patent Application Laid-Open No. 57-160385 will be described as a representative example. Figure 9
Is a position detection circuit diagram showing the configuration of this conventional system,
Following the input terminals a ₃ , b ₃ , c ₃ of each phase of the position detection circuit connected to the three-phase coil, filter circuits a ₄ , b ₄ , c ₄ corresponding to each phase are respectively provided. Each filter circuit has a first-order integration filter with resistors 7 and 8 and capacitors 9 and 10 and a first-order differentiating circuit. Further, in the conventional method, the filter circuit a ₄ neutral voltage combining circuit 12 synthesizes the output voltage of all phases of the to c _4, and the filter circuit a ₄ to c ₄ of the output voltage and neutral-point voltage synthesis It has a rotor position detection circuit composed of comparators a _{5 to} c ₅ for comparing the output voltage of the circuit 12.
Then, the output voltages of the comparators a _{5 to} c ₅ are processed by the logic processing circuit 6 so that the three-phase coil is energized according to the rotor position.

FIG. 10 shows the states of signal voltages at various parts to show this method. Motor phase voltage V _U , V _V , V _W
Is a triangular wave with a π / 2 [rad] phase delayed by an integration filter. ■ V _u dt, ■ V _v dt, ■ V _w d
t. This phase is delayed by ■ V _u dt, ■ V _v dt,
■ create a voltage n obtained by combining the _{V w dt, ■ V u dt} , ■ V v
dt, compares each by the comparator a ₅ to c ₅ and ■ V _w dt and n is _{P U, P V, P W} . This P _U ,
P _V and P _W are input to the logic processing circuit 6, and base signals T _{UH to} T _WL of the motor driving transistors are obtained. The integration filter described above has the characteristics shown in FIG. At this time, when the frequencies of the terminal voltages V _U , V _V , and V _W are the smallest,
That is, by setting the frequency when the number of revolutions of the motor is the minimum to be ωa, it is possible to obtain (1) V _u dt, (3) V _v dt, and (5) V _w dt which always have a phase delay of -π / 2 [rad]. You can do it. However, it is known that the gain of the integrating filter decreases by several dB at the frequency of ωa. This is disadvantageous in that when the rotation speed of the brushless motor is very slow, the value of the terminal voltage of the integration filter is naturally small, and therefore, ■ V _u dt, ■ V _v dt, and ■ V _w dt cannot be detected.

[0005]

SUMMARY OF THE INVENTION Therefore, the present invention provides a rotor position detecting device for a brushless motor, which is capable of reliably detecting the rotor position even when the brushless motor is at a low speed and operating the brushless motor. This is the first issue. The second problem is that the rotor position is accurately detected even if the rotation speed of the brushless motor is high,
An object of the present invention is to provide a rotor position detection device in which the timing of energizing the brushless motor is not delayed.

[0006]

In order to solve the above-mentioned first problem, the present invention provides a rotor position detecting device for a brushless motor, which detects the rotor position by using the neutral point voltage of the brushless motor. A pulse generating means for generating a pulse output having a duty ratio of 50% higher than the frequency of the neutral point voltage, and a pulse width signal having a pulse width between the rising edge and the falling edge of the clock signal using the pulse output as a clock signal. And a first logic circuit for generating a pulse width signal between falling edges delayed by one cycle using the pulse width signal between the falling edges thereof as data and the pulse output of the pulse generating means as a clock signal. And the neutral point voltage as an analog input signal and the pulse width signal between rising edges of the pulse width signal generating means as a logic signal When the analog input signal is "1" and the logic signal becomes "0", the neutral point voltage is gradually changed so as to hold the voltage value when the logic signal becomes "0". Comparing the neutral point voltage with the sample-and-hold circuit that generates the voltage, the neutral point voltage and the neutral point voltage are indefinite in the detection section of the sample-and-hold circuit, and is "1" or "0" in the holding section. Comparing means for generating a stable and unstable unstable pulse output, and using the unstable and stable pulse output of the comparing means as data, the pulse width signal between falling edges delayed by one cycle made by the first logic circuit is used as a clock signal. Second for generating a square wave output having a falling edge and a rising edge which change to "0" and "1" respectively near the positive and negative peak values of the sex point voltage
A rotor position detecting device for a brushless motor is provided. Even when the brushless motor rotates at a very low speed, the neutral point voltage is inevitable and its peak value can be detected.

Further, in order to solve the second problem of the present invention, instead of the pulse generating means, the peak value of the neutral point voltage whose frequency and magnitude change due to the change in the rotation speed of the brushless motor is kept. Peak holding means for generating a voltage output according to the number of rotations of the motor, and voltage controlled oscillation means for oscillating a pulse output with a duty ratio of 50% at a frequency that changes according to the voltage output value of the peak holding means. A rotor position detection device for a brushless motor according to claim 1. Since the output voltage of the peak hold means is a peak value detection signal which always holds the peak value of the neutral point voltage, the output signal and the peak value of the neutral point voltage do not deviate.

[0008]

First Embodiment A first embodiment will be described with reference to the configuration circuit of FIG. 1 and the operation timing charts of FIGS. In FIG. 1, a neutral point voltage V obtained by detecting a difference between a neutral point voltage of the three-phase coil 9 of the brushless motor and a neutral point voltage of three detection resistors R connected in parallel to each three-phase coil 9.
Each of the transistors T _{UH to} T _WH , T of the three-phase inverter circuit 8 for driving the motor is controlled by _MN (hereinafter simply referred to as a neutral point voltage).
It is intended to obtain a base signal of _{UL to} T _WL . E _U First shown in FIG. 2 is the three-phase coils 9, E _V, E _W becomes an induced voltage is generated. In each of these induced voltages, the timing of energizing the brushless motor is at the position indicated by the circle. As a means for detecting the position of this mark, the positive / negative peak value position of the neutral point voltage V _MN is detected. By using the square wave output P _SMN detecting the peak value position of the neutral point voltage V _MN as the input signal of the three-phase ring counter circuit and 7, the bases of the transistors T _{UH to} T _WL of the three-phase inverter circuit 8 are Producing a signal. The detection method of the square wave output P _SMN indicating the positive / negative peak value position of the neutral point voltage V _MN will be described below.

FIG. 3 shows waveforms at various parts regarding the detection method. Duty ratio 50 comprising f _o from the astable oscillator 1
% Pulse output is being generated. The pulse output f _o as the clock signal of the ring counter circuit 2 to obtain a rise between the pulse width signals f _a and falling between the pulse width signal f _b1 of the clock signal. Further, this f _b1 is used as D (data) of the D flip-flop 4, and the pulse output f _o is CK
In addition to the (clock) signal, a pulse width signal between falling edges delayed by one cycle, which is f _b2, is obtained. Following these pulse signals f _o, and the pulse width signal f _a, f _bi, will with respect to the f _b2.

First, when the neutral point voltage V _MN signal of FIG. 3 is used as the analog signal of the sample and hold circuit 3 of FIG. 1 and the rising pulse width signal f _a of the ring counter circuit 2 is _a logic signal, the sample and hold The rising pulse width signal f _a input to the circuit 3 as a logic signal
Is "1"("High"), the voltage value of the neutral point voltage V _MN as an analog input signal is sampled (detected) and stored in a capacitor, and the rising pulse width signal fa as a logic signal is "0". When it becomes (“Low”), the sample (detection) is performed by holding the charge accumulated in the capacitor.
The voltage value at the time of holding is held.

Now, when the stepwise deformed neutral point voltage V _SMN signal and the neutral point voltage signal V _MN obtained here are compared by the comparator 6, the _unsteady stable pulse output P _{VSMN of} FIG. 3 is obtained. The indefinite stable pulse output P _VSMN is the sampling (detection) _period of the sample and hold circuit 3, that is, the rising pulse width signal f _{a of the} ring counter circuit 2 is “1” (Hi
gh) ”, the neutral point voltage V _MN and the step-deformed neutral point voltage V _SMN of the sample-and-hold circuit 3 have the same value, so the indefinite stable pulse output P _VSMN of the comparator 6 is in an indefinite state and may be noise-like. “1” and “0” are repeated at the same time, but the sample and hold circuit 3 holds (holds)
The pulse width signal f _a between the rising edges of the section
Is “0” (“LOW”), the step-deformed neutral point voltage V _SMN of the sample-and-hold circuit is the neutral point voltage V _MN.
On the other hand, since it always shows a large or small value, the indefinite stable pulse output P _VSMN of the comparator 6 is "1" or "0" and a stable output is generated. This stable unstable pulse output P
_{Range VSMN} is obtained which is a period of "1" of falling between pulse width signal f _b1 of the ring counter circuit 2 ( "High"). The falling pulse width signal f _b2 delayed by one cycle _generated by the D flip-flop 4 based on the falling pulse width signal f _b1 is “1” of the falling pulse width signal f _b1.
( "High") for standing climb at (f of _a signal "0" (LOW) center), i.e. in a stable range of "1" or "0" of indefinite stable pulse output P _VSMN of the comparator 61
The pulse width signal f _b2 between the falling _{edges which} are delayed by the cycle rises.

At this time, the indefinite stable pulse output P _VSMN of the comparator 6 is used as D (data) of the D flip-flop 5, and the pulse width output f _b2 between the falling _edges of the D flip-flop 4 delayed by one cycle is output from the D flip-flop 5. CK (clock) and obtain its square wave output P _SMN . This square wave output P _SMN is the indefinite stable pulse output P _{VSMN of the} comparator 6.
Can determine whether the output is stable "1" or "0". That is, when the pulse width signal f _b2 between falling edges delayed by one cycle of the D flip-flop 4 rises, if the indefinite stable pulse output of the comparator 6 is stable "1", the square wave output P _SMN of the D flip-flop 5 is It is “1”, and the pulse width signal f _b2 between the falling edges of the D flip-flop 4 is delayed by one cycle.
When the unstable unstable pulse output of the comparator 6 is stable "0" at the rising edge of, the square wave output P _SMN of the D flip-flop is "0". The rising and falling of the square wave output P _SMN of the D flip-flop 5 changing to “1” or “0” is near the positive and negative peak values of the neutral point voltage V _MN (the pulse of the reference astable oscillator circuit). The higher the frequency of the output f _o , the closer to the peak value of the neutral point voltage V _MN ), which indicates the timing of energization of the brushless motor.

The effect of the first embodiment is that even if the current increases due to an increase in the load on the brushless motor and the neutral point voltage waveform is distorted by the armature reaction, the neutral point voltage is always positive and negative. Since the current can flow to the brushless motor, the motor efficiency is always the best, and even if the motor speed fluctuates, the motor energization timing is high because the current flows at a place where the neutral point voltage is positive or negative. Can be given. This is because the neutral point voltage V _MN always appears even when the brushless motor is rotating at a very low rotation speed, so that its peak value can be detected, and the low rotation speed when the integration filter of the prior art is used. The problem that the rotor position cannot be detected at this time is solved.

[0014]

[Problems of the first embodiment and problems of the second embodiment] The rotor position detecting device of the first embodiment divides the pulse output f _o of the astable oscillation circuit that outputs a certain fixed frequency, and rises it during rising. The pulse width signal f _a is obtained, and the neutral point voltage V _MN is sampled and held by this rising pulse width signal f _a to generate the stepwise modified neutral point voltage V _SMN , which has a magnitude relation with the neutral point voltage. The square wave output P _SMN changing to “1” and “0” in the vicinity of the peak value of the neutral point voltage V _MN is detected by comparing Although the frequency of the pulse output f _o of this time the astable oscillating circuit is fixed for the neutral point voltage V _MN
3 and the square wave output P _SMN produce a deviation α ° of one period of the pulse width signal f _a between rising edges, as shown in FIG.
Frequency f of the neutral point voltage V _MN when the brushless motor is operating at a high rotational (N _HS) is high, the neutral voltage V _MN when the brushless motor is operating at low rotational frequency f ( N _LS ) will be low.

In FIG. 3, when the frequency of the neutral point voltage V _MN is f (N _HS ), the rising pulse width signal f _a input to the sample-and-hold circuit has _a fixed frequency. The difference between the peak value of the point voltage V _MN and the square wave output P _SMN for detecting the peak value is α ° _HS (α ° _HS is an electrical angle of one cycle f (N _HS ) of the neutral point voltage V _MN ). Angle at 360 °). In FIG. 3, the difference between the peak value of the neutral point voltage V _MN and the square wave output P _SMN for detecting the peak value is α.
° _LS (α ° _LS is one cycle f of the neutral point voltage V _MN )
(An angle when (N _LS ) is an electrical angle of 360 °). At this time, whether the rotation of the brushless motor is high or low,
Frequency of rise between pulse width signal f _a to be inputted alpha ° _HS for a fixed, alpha ° _LS time t alpha ° _HS, the t alpha ° _LS is the same.

The frequency of the neutral point voltage V _MN is f
(N _HS) ~f to (N _LS) changes, the time 1 / f
It changes from (N _HS ) to 1 / f (N _LS ). That is, from this, tα ° _HS = tα ° _LS , tα ° _HS / (1 / f
( _NHS ) to 1 / f ( _NLS )). These represent that the ratio of the deviation from the peak value of the neutral point voltage V _MN becomes high as the time of one cycle becomes short (the frequency of the neutral point voltage V _MN is high). That is, the higher the rotational speed of the brushless motor, the later the power supply timing to the brushless motor is delayed, so that there is a problem that the motor efficiency is reduced. The second embodiment aims to solve this problem.

[0017]

[Second Embodiment] FIG. 4 shows the configuration of the second embodiment. The second embodiment is different from the first embodiment in that the astable oscillation circuit is replaced with a peak hold circuit 10 and a voltage controlled oscillator 11. As the rotor position detecting device of the present embodiment, a peak hold circuit 10 for holding the peak value of the neutral point voltage V _MN of the brushless motor three-phase coil 9, and a frequency corresponding to the output voltage V _PH of the peak hold circuit 10. There is a voltage controlled oscillator 11 for oscillating a pulse output f _o,
Further, it includes a ring counter circuit 2, a sample-and-hold circuit 3, first and second D flip-flops 4, 5, a comparator 6, a three-phase ring counter 7, and a three-phase inverter circuit 8.

The second embodiment shows the function of surely detecting the rotor position when the brushless motor is rotating at a low speed as in the case of the first embodiment. The effect is that there is no delay when the energization timing is delayed. Second
The operation and effect of the embodiment will be described below.

FIG. 5 shows the waveform of each part when the brushless motor is rotating at high speed, and FIG. 6 shows the waveform of each part when the brushless motor is rotating at low speed. 7 shows changes in the output voltage V _PH of the peak hold circuit 10 for the neutral point voltage V _MN according to the second embodiment, and FIG. 8 shows the voltage V _PH and the pulse of the voltage controlled oscillator 11 according to the second embodiment. The relationship between the frequency of the output f _o is shown.

The frequency of the neutral point voltage V _MN changes according to the change in the rotation speed of the brushless motor three-phase coil 9. Furthermore,
By frequency neutral point voltage V _MN varies, also changes the magnitude of the neutral point voltage V _MN. At this time, if the peak value of the neutral point voltage V _MN is maintained, the output voltage V _PH corresponding to the rotation speed of the brushless motor can be obtained.
The peak hold circuit 10 functions as an FV converter (FIG. 7). The output voltage V _PH that changes with the change in the rotation speed of the brushless motor is input to the voltage controlled oscillator 11. The voltage controlled oscillator 11 oscillates a pulse output f _o having a duty ratio of 50% with respect to the output voltage V _PH (FIG. 8). Enter the pulse output f _o to the ring counter circuit 2, the rising between pulse width signal f _a, falling between pulse width signal f _b1, further one cycle delay of the pulse width signal f
_b2 is created, the neutral point voltage V _MN is input to the sample and hold circuit 3, and the stepwise modified neutral point voltage V _SMN of the sample and hold circuit 3 is used to obtain a square wave output P _SMN as in the first embodiment. Then, a signal is sent from the three-phase ring counter circuit 7 to the bases of the transistors T _{UH to} T _WL of the three-phase inverter circuit 8.

When the brushless motor three-phase coil 9 is operating at a high speed in FIG. 5, the frequency of the rising pulse width signal f _a of the ring counter circuit 2 is high, and in FIG. 6, the brushless motor three-phase coil 9 is low. When operating by rotation, the pulse width signal f between rising edges of the ring counter circuit 2
Frequency of of _a is low. When the brushless motor three-phase coil 9 is operating at a high rotation speed, the frequency (f (N _HS )) of the neutral point voltage V _MN is high, and the brushless motor three-phase coil 9 is
The frequency (f (N _LS )) of the neutral point voltage V _MN is also low when the engine is operating at low speed. That is, when the frequency of the neutral point voltage V _MN is high, the frequency of the pulse width signal f _a between rising edges which is the input to the sample and hold circuit 3 is increased, and when the frequency of the neutral point voltage V _MN is low, the sample and hold is performed. The frequency of the rising pulse width signal f _a , which is an input to the circuit 3, is lowered. At this time, the difference between the peak value of the neutral point voltage V _MN and the square wave output P _SMN which is the peak value detection signal (FIG. 5α ° _HS , FIG. 6α ° _LS ) changes the frequency of the neutral point voltage V _MN. However, the pulse output f _{o of the} voltage controlled oscillator 11 is always delayed by the same angle with respect to one cycle of 360 °.
To set.

The above operation is a problem of the prior art in the present invention (the neutral position voltage V _MN always appears even if the brushless motor speed is very low, so that the rotor position cannot be detected when the integral filter is used. However, even if the brushless motor speed fluctuates, the energization position to the motor is always at the same position as the rotor position. It is possible to eliminate a decrease in efficiency due to an increase in the number of rotations.

[0023]

As described above, the first invention is a rotor position detecting device for a brushless motor which detects the rotor position by using the neutral point voltage of the brushless motor. High frequency duty ratio 50
% Pulse output means, a pulse width signal generation means for generating a pulse width signal having a pulse width between rising edges and falling edges of the clock signal using the pulse output as a clock signal, and its falling edge. A first logic circuit for generating a pulse width signal between falling edges delayed by one cycle using a pulse width signal between them as data and a pulse output of the pulse generating means as a clock signal, and the neutral point voltage as an analog input signal. The logic signal is "1" with the pulse width signal between rising edges of the pulse width signal generating means as a logic signal.
At this time, an analog input signal is detected, and when the logic signal becomes "0", a stepwise modified neutral point voltage is generated in which the neutral point voltage is changed stepwise so as to hold the voltage value at the time of detection. The sample-and-hold circuit is compared with the step-deformed neutral point voltage and the neutral point voltage, and the sample-and-hold circuit is in an indefinite state in the detection section and stable indefinite at "1" or "0" in the holding section. Comparing means for generating a stable pulse output, and the neutral point voltage using as a clock signal a pulse width signal between falling edges delayed by one cycle made by the first logic circuit using the indefinite stable pulse output of the comparing means as data. The brushless motor includes a second logic circuit that generates a square wave output having a falling edge and a rising edge that change to "0" and "1" in the vicinity of the positive and negative peak values, respectively. Even rotating perform reliably rotor position detection, there is an excellent effect that it is possible to operate the brushless motor.

In the second aspect of the invention, instead of the pulse generating means, the peak value of the neutral point voltage whose frequency and magnitude change due to the change in the rotation speed of the brushless motor is maintained and is kept depending on the rotation speed of the motor. When the brushless motor rotates at a low speed, the peakless means for generating a voltage output and the voltage controlled oscillating means for oscillating a pulse output with a duty ratio of 50% at a frequency that changes according to the voltage output value of the peak hold means are provided. There is an excellent effect that the rotor position can be surely detected, and the rotor position is accurately detected even when the brushless motor is rotated at a high speed, so that the timing of energizing the brushless motor is not delayed.

[Brief description of drawings]

FIG. 1 is a circuit diagram showing a configuration of a rotor position detecting device according to a first embodiment of the present invention.

FIG. 2 is a timing chart showing the operating principle of the first embodiment.

FIG. 3 is a timing chart showing an actual operation of the first embodiment.

FIG. 4 is a circuit diagram showing a configuration of a rotor position detecting device according to a second embodiment.

FIG. 5 is a timing chart showing waveforms of various parts of the device of the second embodiment when the brushless motor is rotating at high speed.

FIG. 6 is a timing chart showing waveforms of various parts of the device of the second embodiment when the brushless motor is rotating at a low speed.

FIG. 7 is a waveform diagram showing changes in the output voltage of the neutral point voltage peak hold circuit according to the second embodiment.

FIG. 8 is a characteristic diagram showing the relationship between the output voltage of the peak hold circuit and the pulse output frequency of the voltage controlled oscillator according to the second embodiment.

FIG. 9 is a circuit diagram showing a configuration of a conventional rotor position detecting device.

FIG. 10 is a timing chart showing states of signal voltages of respective parts in the conventional method.

FIG. 11 is a characteristic diagram showing characteristics of an integration filter used in a conventional method.

[Explanation of symbols]

1 ... Astable oscillator circuit, 2 ... Ring counter circuit,
3 ... Sample and hold circuit, 4, 5 ... D flip-flop, 6 ... Comparator, 7 ... Three-phase ring counter circuit, 8 ... Three-phase inverter circuit, 9 ...
Brushless motor three-phase coil, 10 ... peak hold circuit, 11 ... voltage controlled oscillator, 12 ... comparator. f _o ... Pulse output, f _a ... rising between pulse width signal, f _b1 ... falling between pulse width signal, falling between the pulse width signal f _b2 ... 1 cycle delay, V _SMN. .. Step deformation Neutral point voltage, P _VSMN ... _Unstable pulse output, P _SMN ... Square wave output, V _MN ... Neutral point voltage, R ... Detection resistor.

Claims (2)

[Claims] 1. A rotor position detecting device for a brushless motor for detecting a rotor position by using a neutral point voltage of the brushless motor, wherein a pulse output having a duty ratio of 50% having a frequency higher than the frequency of the neutral point voltage is provided. A pulse generating means for generating, a pulse width signal generating means for generating a pulse width signal having a pulse width between rising edges and falling edges of the clock signal using the pulse output as a clock signal, and a pulse width signal between the falling edges. Is used as data and a pulse output of the pulse generating means is used as a clock signal to generate a pulse width signal between falling edges delayed by one cycle, and the neutral point voltage is used as an analog input signal to generate the pulse width signal generating means. The rising edge pulse width signal is used as a logic signal, and when the logic signal is "1", the analog input signal is detected and the logic signal is " When it becomes ", the sample-and-hold circuit that generates the step-deformed neutral point voltage in a form of gradually changing the neutral point voltage so as to hold the detected voltage value, and the step-deformed neutral point voltage And a neutral point voltage, and a comparator means for generating a stable and unstable pulse output at "1" or "0" in an undefined state in the detection section of the sample-and-hold circuit and a comparison section. Is used as the data, and the pulse width signal between the falling edges delayed by one cycle generated by the first logic circuit is used as the clock signal, which are "0" and "" respectively near the positive and negative peak values of the neutral point voltage. And a second logic circuit that generates a square wave output having a falling edge and a rising edge that change to "1". 2. In place of the pulse generating means, the peak value of the neutral point voltage whose frequency and magnitude change due to a change in the rotation speed of the brushless motor is maintained and a voltage output according to the rotation speed of the motor is generated. 2. A rotor position detecting device for a brushless motor according to claim 1, further comprising: peak-holding means for activating the voltage, and voltage-controlled oscillating means for oscillating a pulse output with a duty ratio of 50% at a frequency that changes according to the voltage output value of the peak-holding means. ..