JPS63253892A - Brushless motor - Google Patents

Abstract

PURPOSE:To lower the cost of a device and improve the quality, by analyzing the influence of an FG pulse number upon a system, and by using a sensor in common in the range of the influence. CONSTITUTION:A rotor magnet 2 and a speed detecting magnet 3 are supported by the rotor 1 of a brushless motor, and the rotor 1 is connected to a shaft 4 and is supported by a stator frame 7 via bearings 5-6. Its rotational force is generated by a stator core 8 and a coil 9 which are supported by the frame 7, and said rotor magnet 2. The magnet pole detecting sensor 11 of the magnet 2 for switching the exciting timing of the coil 9 is set near the magnet 2, and a speed detecting coil 12 is set near the magnet 3. Then, respective sensor signals are fed to the coil exciting phase switching signal circuit of a motor driving circuit, and the like, and by an FG pulse number, the sample periods of speed feedback and phase feedback are set, and a factor of a gain is set, and specified speed/position control is performed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a brushless mode that performs constant speed control, and particularly to synchronization control of a laser scanning scanner mode of a laser beam printer (LBP) that requires high precision.

[Conventional technology]

Conventionally, there have been various methods for constant speed control of DC motors, and in recent years, synchronization control technology has been particularly widely used. An example of this is 7 of ``Control Circuit Design of rDC Motor'', written by Kinji Tanikoshi, published by CQ Publishing Co., Ltd., May 1985.
It can be seen on page 1, etc., and the principle is quoted in Figure 22.
FG3, a two-phase speed sensor attached to the motor 308
The signal of 09 is fed back to the phase comparison circuit 301 via the amplification and shaping circuit 305, where it is phase-compared with the signal from the reference oscillation circuit 300, and the phase difference is sent to the low-pass filter 302 of the phase difference-voltage conversion circuit. is input to the power amplification circuit 3, and is controlled with a voltage according to the phase difference between the reference signal and the power amplification circuit 3.
Drive 07. In addition to this, the speed feedback is input into the F-V conversion circuit 304 and compared with the base plate 303, and according to the speed difference, a voltage is outputted to the mixing circuit 306, which is added and mixed with the output of the above-mentioned low-pass filter 302. We aim to further stabilize servo control. Examples of these actual circuits were created in November 1982.
It can be found on page 120 of "rDC Brushless Motor and Control Circuit" written by Kinji Tanikoshi and published by Tsuki Sogo Electronic Publishing Co., Ltd., and is quoted and shown in FIG. In contrast to FIG. 22, the reference oscillation circuit 300 uses a crystal, and the speed reference 30
The six-pronged phase comparator circuit 301. Low-pass filter 302, F-V conversion circuit 304. The widening and shaping circuit 305 collectively forms an integrated circuit 310.

A mixing circuit 306 connects the output of the low-pass filter 302 and the output of the F-V conversion circuit 304 to a buffer amplifier 31, respectively.
1,312, and is mixed in an operational amplifier 314. The power amplification circuit is composed of transistors 315 and 316. and.

The FG 309 uses a photo encoder in which a slit disk 320 rotates between a light emitting element 321 and a light receiving element 322.

As is customary in this type of control, as disclosed on page 125 of Publication Example 2, it is said that as many FG pulses as possible is good for control. Then, an encoder 309 as shown in FIG. 23 is used.

The present invention provides a technique for achieving the purpose at low cost by using this FG in common with other sensors without installing it specially.

[Problem that the invention seeks to solve]

How can the FG disclosed in the above prior art be used in common with other sensors without special installation, and how to deal with the reduction in pulses and increased noise that are limited by the shared use, and constant speed control? The problem is how to ensure the stability of

[Means for solving problems]

The above purpose is to analyze the influence of the number of FG pulses on the system, determine the limit of the required number of pulses, and realize sensor sharing within that range. Analyze the range that does not have a negative impact on
The purpose is to satisfy the conditions and determine the constant configuration of the system.

[Effect]

The number of FG pulses determines the sample period of velocity feedback and phase feedback, and is also a factor in the gain of phase feedback. It is possible to cover the latter gain factor with other in-circuit amplifiers.

Therefore, the former sampling period can reduce the number of pulses to a level that does not cause problems for the system. Further, the noise contained in the feedback signal can be removed by a filter within a range where the delay time of the filter does not disturb the stability conditions of the system.

[Embodiments of the invention]

Hereinafter, the principle of the present invention will be explained with reference to the drawings.

FIG. 1 shows a cross-sectional structure of an example of a brushless motor to which the present invention is applied. The rotor 1 is coupled to a shaft 4 while supporting a rotor magnet 2 and a speed detection magnet 3, and is rotatable relative to a stator frame 7 via bearings 5 and 6. The rotational force is applied to the stator core 8. which is supported by the stator frame 7. This occurs based on the well-known principle of a motor constituted by the stator coil 9 and the rotor magnet 2. A rotating magnetic pole detection sensor (
For example, a Hall element) 11 is installed near the magnet 2, and a speed detection coil 12 is installed near the speed detection magnet 3, but they are attached to a wiring printed circuit board 10 supported by the stator frame 7. Or formed.

Fig. 1 shows a type with a radial gap and a stator core 8, but the example shown in Fig. 2 has an axial gap and does not have a stator core 8, and also does not have a speed detection magnet 3. Rotor magnet 22 by type
This is a type that detects speed using magnetic force, such as multifaceted mirror 2.
This is a type of scanner motor for laser printers with 5.

Rotor 21. Axis 24. Bearings 25, 26. Stator frame 27. York iron plate 28. Stator coil 29. Printed circuit board 30. It has a rotor magnetic pole detection sensor 31 and rotates on the same principle as the example in FIG.

FIG. 3 is a block diagram showing the principle of coil excitation phase switching operation of the brushless motor of FIGS. 1 and 2, and FIG. 4 is a timing chart thereof.

The motor 41 has a stator coil U phase 42. In this example, the rotor magnetic pole position detection Hall element group 45 corresponds to a three-phase example having a V-phase 43 and a W-phase 44.
47.48, the detection signal of the Hall element group 45 passes through the Hall element signal amplification and waveform shaping circuit 52 of the brushless motor drive circuit 50, and enters the phase switching control logic circuit 53.
Then, the conduction of the output amplifier circuit 54 forming a bridge arm etc. corresponding to each phase coil is controlled, and as a result,
As shown in FIG. 4, the coils of each phase are excited and generate rotational force as a motor.

In FIG. 4, Su and Sv of the rotor magnetic pole detection signal 60.

SW rises with a difference of 60" in electrical angle, and the duty is 180@. From these signals, circuit 5
3.54 control, the motor coil excitation voltage 70 is set as 120@conducting and 60@non-conducting with a 120 degree difference between the phases in the aftereffect of three phases such as Vu, VV, and VW for each phase.

This type of motor is effectively utilized in the feedback control phase as shown by the block in FIG. The motor 41 has a rotor magnetic pole detection sensor 45. It has a speed detection sensor 832 and a position detection sensor 84, and the signals from each sensor are sent to the coil excitation phase switching signal circuit 51. of the brushless motor drive circuit 5o. It is coupled to the feedback signal control circuit 82 of the speed position control circuit 81, and electrical energy from the power source 80 is supplied to the motor 41 through the power control circuit 85.54 while being controlled, thereby realizing predetermined speed and two position control. .

The description based on the self-a control logic of speed-2-position control is first explained in the sixth section.
It is expressed as a model as shown in the figure. The motor 41 has an inertia stone and a control Do, and is operated by a servo amplifier 100 with a gain G including a torque constant and a current drive source, and its speed is detected by a detection sensor 1019, and its position is detected by a sensor 102, and ω and θ are detected, respectively. is fed back as, each standard human power ω
1. θ, and the deviations ωE and θE are summed up via the gains Gω and GO of the feedback section 97.92, and the process reaches block 100. At this time, the load disturbance d is caused by block 9
The signals are summed together via a gain G of 8.

The equation of motion at this time is as follows.

...Equation (1) However, the inertia of the Jo" rotation axis (g, B, S2) D = Gw-
Gwa+Do (g, am/rad/S)...Formula (
2) % formula %) ...Formula (3-1) F ” Ga a・G, ...Formula (3-2
) It can be seen that K is the synchronizing force and D is the tamping force.

Expressing these in the form of a Laplace-transformed transfer function and focusing on the reference input OK and the disturbance d(g, am), we get:
As shown in FIG. 7, each overall transfer function is in block 11.
It becomes something like 0,111. The characteristic equations are as follows: JS2+DS+=O-...Formula (4) It can be seen that the servo characteristics of quick dissolution and stability can be considered by considering the reference input θ amount and the disturbance d in the same way. The model is shown in Figure 8, and is a second-order vibration system known to all. The resonance frequency and damping coefficient are as follows. The transfer function of block 112 is as follows.

Brushless motors can be effectively used in all fields of OA due to their various advantages. There are many advantages to performing phase synchronization control, especially for constant speed control operation. In the phase synchronization control, the position θ shown in FIGS. 5 and 6 is the rotation angle of the rotor with reference to the synchronization coordinates, and the braking Do is the control coefficient of the motor at the synchronization speed. It can be treated similarly in FIG. θ can be recognized from the concept of load angle of a synchronous motor, while Do
In practice, the 0 phase synchronization control is small and can be ignored.

It is advantageous to perform speed feedback control digitally as well, which can be expressed as shown in FIG. Reference oscillation circuit 130
, the speed reference ω12 phase reference θ fish is emitted. The phase synchronization control circuit 120 consists of a 1-phase control circuit 121 and a speed control circuit 124, each of which has a gain Go as shown in FIG.
, Gω, which detects the rotation of the motor 41 and sends feedback signals from a frequency generator sensor 131 such as an encoder that feeds back signals of speed and rotor position, that is, phase, to phase bones θ1 . Receive the velocity component ω, base $0. Compare with ω. And the circuit 121 is (θ
Mu θ) phase comparator circuit 122 and 0ε=(θ1-0)
It has a phase-voltage conversion circuit 123 that generates a voltage corresponding to , and an addition corresponding section 127 of gain go. Further, the circuit 124 includes a frequency comparison circuit 125 of (ω1-ω) and a frequency comparison circuit 125 of (ω1-ω).
It has a frequency-voltage conversion circuit 126 that generates a voltage corresponding to E=(ω, -ω) and an addition corresponding section 128 of gain go. The respective outputs are shown as VO and VO in FIG. 9, and an explanatory example thereof is shown in FIGS. 10 and 11 with their sizes normalized.

Further, VO and VO are added together as in the cylinder 12 described later, and also passed through a filter 129.

In FIG. 10, the phase synchronization range is the gain G
It is an element included in ω and is involved in the jump of the system into the saturation region described below. In other words, outside the samurai range is VO and V in Figure 11.
e' is 100% (low speed side) or 0% (high speed side).

θ=100% in Fig. 11 is the electrical angle 2 rc / (ra
2π/FG (ra
d) is the mechanical angle. District 0 is half that amount.

Further, 100% saturation of VO1 and Vθ' is approximately equal to the constant voltage of the control circuit.

When such a phase synchronization control system is expressed in terms of automatic control theory, it also includes a part that is a discrete value system. That is, the detection of speed ω and phase θ by the sensor 131 takes one period T = 2π
/ω(S), and on the other hand, the outputs ■θ and Vω of the phase synchronization control circuit 120 are zero-order hold circuit outputs that have been digital-to-analog converted. Therefore, the Z-transformation method.

Subject to pulse transfer function techniques. As another nonlinear factor, as explained in FIGS. 10 and 11, the values of ω' and vo' have a saturation region. Furthermore, many systems such as the power supply 80 shown in FIG. 5 do not have a function of braking the motor 41 against overspeed. However, the sample value of the sensor and the Vθ of one circuit output.

With respect to the zero-order hold property of Vω, by treating the period T as a dead time, and only within the phase synchronization range, practical design and improvement considerations can be made, and furthermore, the dead time can be ignored. However, the basics of servo characteristics can be considered.

In that case, the theory will be based on FIGS. 6, 7, and 8. Compared to FIG. 6, since ωL is given together with 01 from the circuit 130 as shown in FIG. 9, ω,=Sθi
... Formula (8) is obtained, and the numerator of block 110 in FIG. 7 is as follows.

DS+K...Equation (9) Other aspects remain unchanged.

Furthermore, the outputs Va, V of the phase synchronization control system 120 in FIG.
When ω is passed through a filter with a time constant Tf of 1, it is the denominator of blocks 110 and 111 in FIG.

That is, the characteristic equation of formula (4) is as follows.

ZS"+JoS"+DS+=O+・+Equation (1o) However, Z=TtJo (g-cyn-33) −・Equation (11) And the operation stability conditions according to the Hurwitz method in automatic control theory are as follows.

JoD-7K>O...Formula (12) The molecule of block 111 in FIG. 7 is as follows.

F(1+TzS)...Formula (13)
To describe in more detail, each constant is as follows.

G, = Ga-KT/to 1 (g-am/v)...Formula (]
4) % formula % (including filter) KT = Torque constant of motor 41 (g-1/A) Km =
Current feedback gain (Ω) to the servo amplifier section Ga Δ ω where Vc = voltage equivalent to 100% of Vω in Figure 10 (V) Δω = phase synchronization range (rad) in Figure 10 gω = 9th
Gain 2π of block 127 in the figure. However, Fa = Number of pulses of sensor 131 per revolution of motor 41 that effectively act on circuit 121 Vc = Voltage equivalent to 100% of Vθ in Figure 11 (V) go = Voltage (V) in Figure 9 Gain of block 128 Even if described in detail in this way, FIGS. 6-8.

Alternatively, the basics of the servo characteristics in equations (10) and (12) do not change, however, in the phase synchronization control system, the reference input OL and ωl are constant, and the servo characteristics are within and near the synchronization speed. A practical method is to consider the load disturbance d(g-am) as shown in block 111 in FIG.

In order to realize the servo characteristics by such handling, the following conditions are specifically required in correspondence with the above-mentioned conditions.

First of all, in order to approximate the system to a continuous system and achieve performance without leaving it in the range of a discrete value system, it is desirable that the sampling period T be 1/1o or less of the mechanical constant T of the system. T1 and T are as follows.

EKT However, R = Coil resistance of motor 41 (Ω) Jo = Inertia of rotating shaft (g-ro・52) KE = Back electromotive force constant of motor 41 V / rad /
KT = torque constant of the motor 41 (g-am/A), where N = target constant rotation speed (rpm) of the motor 41 ω = angular velocity corresponding to the above N (rad/S) FG = sensor 131
The number of pulses per rotation of the motor 41 that actually acts on the circuit 121 is therefore the condition: T1・NT,・ω, where the ratio is γ.In other words, the number of pulses of the sensor 131 is equal to the target speed and the time of the motor 41. It needs to be determined by equation (19) in inverse proportion to the constant. To give an example, in the example shown in Figure 1, T,
=0.04 (S), N=370 (rpm).

γ=11 is achieved with FG=45. In addition, in the example of the motor shown in FIG. 2, T=0.45 (S).

N = 7700 (rpm) From equation (19), F
C>0.17. 1 pulse per rotation, ie FG=1
Even if.

γ=57.8. This suggests that in FIG. 2, the signal from the rotor magnetic pole detection sensor 31 can be used even if the speed detection coil 32 is not provided. This means that a part of the signal 60 in FIG. 3 or 4 is used, and in FIG. 5, all of the sensors 83 and 84 are replaced by the sensor 45.

However, in reality, sensor 45, sensor 83.84
One problem with substituting is jitter. In other words, if you explain it with Figure 2 etc., it is a rotor magnet!・22
There are variations in the pitch of the magnetic poles, and the rotor magnetic pole detection Hall element 31 has variations in sensitivity, mounting position, etc. As a result, each of the signals 60 in FIG. 4 has jitter, and each signal Su , Sv, and Sw, no matter how they are used in combination, they basically have ratio jitter, which ranges from several percent to several tens of percent. On the other hand, the output characteristics of the circuits 121 and 124 in FIG. 9, which operate in response to the signal from the sensor 131, are as follows.
As shown in FIGS. 10 and 11, it can be understood that the jitter of the sensor 131 becomes noise on the horizontal axis ω, especially as shown in FIG. is the relative value of Δω is selected to be around ±4% or more, taking into account the characteristics of synchronization pull-in and synchronization maintenance. Therefore, the jitter of the sensor 131.

This becomes the same level as the magnitude of Δω, and the signal-to-noise ratio becomes large, making it difficult to realize servo characteristics such as system stability. The same applies to 0 in FIG. To improve this, it is effective to insert a low-pass filter into the circuits 127 and 128 in FIG. 9, or after them. The filter temporary constant at that time becomes the second condition.

That is, the equations (10), (11), and (12) are as described, and the detailed breakdown is as shown in equations (2), (3-1), and (1).
4) Substituting and rearranging , (15), (16), etc., from equation (12), however, Vc in equation (15) and Vc in equation (16) are the same 6 In other words, the filter temporary constant is determined by the sensor 131. The larger the phase synchronization range (inversely proportional to them)
Block 127, 1 for addition of phase, velocity, etc.
It can be seen that it depends on the gain ratio of 28.

An example of the actual configuration of TI, -gω9gθ is shown in FIG. 12, proceeding from FIG. Phase-voltage conversion circuit 123° frequency-voltage conversion circuit 126 Hold outputs Vθ' and ■ω' from each digital-to-analog conversion circuit are resistors Rf),
The signals are added by the operational amplifier 133 via Rω. The operational amplifier 133 is also connected to resistors Ra and Ram Ra as shown in the figure, as well as a filter capacitor C (F) 132, and the output eo is as follows.

=gθvo'+gωVω'...Formula (22)
However, the resistor Ra can be selected or eliminated as necessary.

Then, the filter temporary constant Tf is as follows.

Tt=R-c (s)...Formula (25
) The circuit as shown in FIG. 12 described above is expressed as 1, filter block 129 in FIG.

Again, the circuit 120 of FIG. 9 corresponds to the circuit 82 of FIG.

13 and 14 show an example in which the rotor magnetic pole detection sensor 45 is substituted for the sensor 131.

In FIG. 13, the rotor phase signal from the sensor group 45 is passed through a signal AC coupling capacitor group 135° amplification, a waveform shaping circuit 136, and a frequency generator 137 that operates at the zero crossing, rising edge, or falling edge of the waveform. The phase and velocity feedback signals θ and ω are generated through the sensor 131 and serve as a substitute for the sensor 131. FIG. 14 shows one circuit 136.

This is an example in which the circuit 52 in FIG. 3 is used in common. If the pulse number FG satisfies the conditions in both FIGS. 13 and 14, it is also possible to use only the individual sensor 46, 47, or 48 in the sensor group 45 to simplify the subsequent circuit.

An actual example of the power control circuit 85 shown in FIG. 5 is shown in FIGS. 15 and 16 with a current fee back.

In FIG. 15, 1 from the phase synchronization control circuit 120
The phase and speed feedback voltage eo and the voltage across the detection resistor 152 (negative feedback) of the current i flowing through the moder 41 are added by an operational amplifier 153.

A current drive source type analog control servo amplifier that controls the amount of conduction of the transistor 151 and controls the current conduction of the power transistor 150 is formed.

Figure 16 shows that the analog method is stopped and the pulse width control (PW
In this example, the output of the arithmetic amplification amplifier 153 is compared with the triangular wave of the constant period triangular wave oscillation circuit 155 via the inverter 154, the comparator 156 generates a duty control rectangular wave, and the transistor 151
This is an example in which the power transistor 150 is controlled by PWM.

In any case, the gain in this part is practically
This corresponds to the gain G of block 100 in FIG. 6 and the gain G of block 100 in FIG. 9, which are summarized in the torque constant KT (gaI/A) of the motor 41 and the current feedback gain ki.

When the above-described components are combined as a whole to configure a system, the result will be as shown in FIG. 17. Figure 13.

The area around the sensor as shown in Fig. 14 is the sensor 13 in Fig. 9.
The sensor 83.1 is described corresponding to 1 and is referred to in FIG.
84 is the form substituted by the sensor 45. That is, formulas (18) and (19) are satisfied.

FIG. 18 shows an example of actual operation data in this configuration.

(a), (b), and (c) all show the time changes of the phase output Vθ', speed output Vω', and hemi-torrent current i before and after synchronization pull-in after startup, but conditions such as stabilization are required. Become. (
a) is a state without the filter 129 in FIG.
■Even after the phase θ expressed by θ′ has stabilized, a ripple remains in the current i. This means that the speed voltage Vω' is the sensor 131
This is due to the presence of ripples due to jitter. Therefore, the ripple in the current i not only causes a torque ripple, but also has a subtle negative effect on the stability of the entire system. (b) is the result of adding only the filter 129 as shown in equation (25) as an improvement to (a), that is, adding a capacitor (132) to FIG. ) is violated, indicating that the system is diverging. However, the current i is omitted. (c) shows this improvement in circuit 127 of FIG.
, 128 phase feedback go velocity feedback gω, that is, by changing RO and Rω in FIG. 12 as an example,
Conditional expressions (12) and (20) % Since the jitter of the sensor 131 is basically unchanged, the ripple of VO2 is not improved, but the ripple of the current i is greatly improved, and the phase is 0.■ It can be seen that the response of θ' is also improved.

In this way, the negative influence on the system of the filter 129, which compensates for the disadvantage of jitter of the simplified and shared sensor 131, can be reduced by using the phase and speed feedback gain go9.
Good characteristics can be obtained by compensating for the balance of gω.

In addition to the above description, the range can be further analyzed as follows, including experimental information.

First, if we transform Equation 1 (22), we get... Equation (26) There is an output limit, that is, saturation, depending on the power supply voltage value for the feedback voltage eo and its subsequent stages, and within that linear range, the voltage vO'vω' However, as shown in FIGS. 10 and 11, the phase is 0. I would like to think of it as changing effectively by feedback of the speed ω and having an effect on eQ. At this time, since the speed ω is the change in the phase θ, it is assumed that the saturation value is the same for both, and it is assumed to be 1.
Also, let the change in θ be sin αt.

However, eo' = change in eo This change is minimized by considering the effective use of the linear region of the system.

Kottoki e o' = 1 ~ 1, 4 (= E)
t 6° On the other hand, the disturbance (load) d (g-a
Considering the steady-state deviation for ll), applying the final value theorem of automatic control theory to FIG. 7, we get the following.

θ=θs + -d (rad) Complete set (29), °, Load torque=Fd=K CO-01)
C0-01) (...Equation (30)) And K is from Equations (3-1), (16), and (23). This is the change in equation (28) after receiving a bias.

The bias at this time is smaller as the resistance go is larger, that is, as Rθ is smaller. Therefore, it is suitable for high loads and g.

Figure 19 shows the gain go after taking some load torque into consideration.
An example of the experimental results of the stable operation region with respect to changes in 2gω is shown. The area was 200 surrounded by a dashed line. The axis scale is normalized, and □g ω % is the actual resistance value IMΩ.

When the conditions are shifted from the stable region to the six directions, the phase θ becomes oscillatory, and the opposite B direction becomes the overflow direction, which is consistent with the theory described above. Furthermore, the condition in the C direction causes ω to approach the low speed side in Figure 10, and the output saturation of It can be confirmed that the vibration is induced. From these experimental analyzes it can be determined that g ω is within the practical range.

In order to improve the characteristics of servo systems, in addition to gain, phase delay and phase lead are also known as general techniques.

There are various compensation methods for feedback, but the present invention proposes a simple and basic method based on theoretical analysis and experimental results, before making the system more complicated and increasing the cost by compensating for them more than necessary. This is what we are trying to achieve. Naturally, the ripple countermeasure filter 12 of the sensor 131 shown in FIG.
By inserting 9, the system becomes higher-order, and formulas (12) and (2
0) is no longer satisfied or a similar unstable region exists, it is clear that it is not a good idea to rely on phase compensation or additional feedback compensation without considering the balance of the feedback gain gθ9gω. It can also be seen that these are within the scope of the present invention.

In configuration, the filter 129 is not limited to the connection such as the capacitor 132 shown in FIG.
In addition, it is also practical to connect an operational amplifier of a voltage follower circuit between the circuit 123 and RO and between the circuit 126 and Rω to separate the impedance between the circuits. However, all of these are within the techniques of the invention described above.

In such a case, a filter can be inserted into each of the phase and velocity feedbacks to eliminate the adverse effects of ripples. An actual example is shown in FIG. 20, and a corresponding servo block diagram is shown in FIG. 21.

In FIG. 20, voltage follower circuit 202.2
03, 212, 213 are the phase-voltage conversion circuit 1 as shown in the figure.
239 frequency-voltage conversion circuit 126,
Resistors 204, 214, respectively, in between. capacitor 205
, 215, forming a filter with a time constant determined from the constants. The subsequent stage is the same as that shown in FIG.

In addition, the handling of the gain gO + gω is as follows: resistance 204°21
As a result, as shown in FIG. 21, the filters 207 and 217 with time constants Tz and
7, and the filter 129 is described after that. In this case, the characteristic equation of the system is made higher order than the third order of equation (10) as shown below.

(TzTzTa) JoS5+ (TzTz+TzTa
+T3Tz) JoS'+ (T1+T2+T3) J
O8”+ (JoTzD) S”+ (D+TaK)S
+ = O... Equation (35) Stability discrimination of this system can be studied using the well-known theory of stability discrimination, but in practice, filter 1
29, 207. .

217 If you can consider one of them and leave out the others, the cost will be low. Only the filter 129 has already been mentioned.

Assuming only the filter 207 and nothing else, the formula % formula %...Equation (36) The stability condition by the Hurwitz method is actually designed to satisfy K > J o, so it is a practical condition.

If only filter 217 is used, T□=O in equation 1 (35)
, Tz = O, T a J o S 3+ J o S ” + (D
+ T a K ) S + K = 0...Formula (
39) The stability conditions according to the Fulvic method are as follows.

D>O...Equation (40) It is obvious that these conditions apply to the system expressed in FIG. 21 even if the system is not configured like the example in FIG. 20 itself.

〔Effect of the invention〕

As described above, according to the present invention, 1. Since the second sensor can be shared, the servo system can be simplified, the range of applications can be expanded, and a high-quality brushless motor can be provided at a lower cost.

[Brief explanation of drawings]

Figure 1 is an example of a cross-sectional view of a conventional brushless motor. FIG. 2 is a diagram showing another example similar to FIG. 1, and FIG. 3 is a diagram explaining the conventional principle of coil excitation phase switching of a brushless motor.
Fig. 4 is a time sequence corresponding to Fig. 3, Fig. 5 is a block diagram explaining the system configuration of the present invention, Fig. 6 is a servo block model diagram corresponding to Fig. 5, and Fig. 7 is a transmission corresponding to Fig. 6. Function model diagram, Figure 8 is a secondary vibration system model expression diagram corresponding to Figure 7, Figure 9 is a block diagram of the configuration principle of the present invention, Figure 1
Figure 0 shows an example of the output characteristics of a frequency-voltage conversion circuit, Figure 11 shows an example of the output characteristics of a phase-voltage conversion circuit, Figure 12 shows the addition of phase feedback voltage and speed feedback voltage, and the configuration of a sensor ripple removal filter. 14 and 14 are examples of the configuration of a part of the sensor, FIGS. 15 and 16 are examples of the configuration of the servo amplifier section, FIG. 17 is a system configuration diagram of the present invention, and FIGS. 18 and 19 are the products of the present invention. 20 is an example of another configuration of part of the filter, FIG. 21 is a servo block diagram corresponding to FIG. 20, and FIGS. 22 and 23 are diagrams showing conventional circuit configurations. be. 41... Motor, 45... Rotor magnetic pole detection sensor, 50... Brushless modele drive circuit, 80...
power supply. 85... Power control circuit, 120... Phase synchronization control circuit, 130... Reference oscillation circuit.

Claims (1)

[Claims] 1. A brushless motor that includes a first sensor that detects the position of a rotating magnetic pole of a permanent magnet provided in a rotor, and that switches the excitation phase of a stator coil based on the first sensor signal, comprising: A second sensor that detects the phase and speed is provided, and this second sensor signal is compared with a reference signal from a reference oscillation circuit provided separately, and the respective and a digital-to-analog conversion circuit that generates a control voltage, and a phase synchronization circuit that performs constant-speed phase synchronization control using both of them, and the second phase synchronization circuit that is fed back to this phase synchronization circuit.
The number of pulses FG per motor revolution of the sensor satisfies the relationship FG>600/(T_m・N) with the product of the motor's mechanical time constant T_m and target constant speed N, A brushless motor characterized in that the first sensor and the second sensor are shared. 2. The brushless brushless according to claim 1, characterized in that a filter for removing jitter of the sensor is connected to a subsequent stage of a voltage conversion circuit that outputs a voltage according to a phase difference or a speed difference. motor. 3. In the device according to claim 2, the time constant T_f of the filter is equal to the speed feedback gain G_w.
A brushless motor characterized in that the value is smaller than the value obtained by dividing the phase feedback gain Gθ by the phase feedback gain Gθ. 4. The filter according to claim 2, characterized in that the time constant of the filter is smaller than the value obtained by dividing the total moment of inertia J_0 of the rotor by a coefficient D that determines the damping force of the system. brushless motor. 5. In the device described in claim 2, a gain g corresponding to speed feedback at a subsequent stage of the speed difference-voltage conversion circuit.
A brushless motor characterized in that a ratio g_w/gθ of _w and a gain gθ corresponding to phase feedback at a subsequent stage of the phase difference-voltage conversion circuit is smaller than 2. 6. The device according to claim 5, wherein the gains g_w and gθ are determined by the input resistance R_w or Rθ of an adder circuit of an operational amplifier provided after each voltage conversion circuit, R_w/Rθ<
A brushless motor characterized by: 2.