JPH0675637A - Magnetic head positioning control system - Google Patents

Abstract

PURPOSE:To reduce the cost of an encoder by compensating variation in the gain of the encoder and improving follow-up precision. CONSTITUTION:A deviation signal detection part 1 obtains deviation (e) from a head, sector by sector. The deviation (e) is inputted to a deviation compensator 2. A memory part 3 is stored with the previously measured eccentricity r* of media by learning. Further, a position signal detection part 9 obtains a motor position x* detected by the encoder. Then an encoder gain compensator 4 sequentially calculate the displacement-output gain Ke of the encoder and a correction coefficient He being its reciprocal from the detected motor position x* of the head, the predicted eccentricity r* of the media, and the motor position x* detected by the encoder. Then the correction coefficient He is rewritten each time the detected deviation (e) of the head is read out. Then a multiplier 6 multiplies the motor position x* detected by the encoder by the correction coefficient He.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a magnetic head positioning control system for a magnetic recording / reproducing apparatus using a sector servo system.

[0002]

2. Description of the Related Art A step motor can be easily positioned in an open loop, which is advantageous in terms of cost. Therefore, it is widely used for rough positioning applications, for example, OA equipment such as printers, plotters, and disk devices.

However, a closed loop servo system using a DC motor and an encoder is often used for minute positioning. Further, a method may be considered in which the stepping motor is micro-step-driven and the positioning point is interpolated to improve the positioning accuracy. Therefore,
There is a control drive system using a microstep motor and an encoder. The microstepping motor is a stepping motor whose linearity is improved satisfactorily by suppressing the cogging torque to a minimum by modifying the conventional magnetic circuit.

In the closed loop servo system of the step motor, there are a system for controlling the amount of current and a system for controlling the phase angle. The phase control system will be described below. With the phase control method, the position of the rotor is detected by the encoder,
This is a method of positioning the rotor by controlling the magnetomotive force phase of the stator with respect to the rotor.

FIG. 3 shows the configuration of this control device. In the figure, the output of the deviation compensator 21 is input to the limiter 22. The output of the limiter 22 becomes the electrical angle h, and the function generator 23 receives the signal as the electrical angle e. The output of the function generator 23 is the PWM 25
To the coils 26a, 2
It is input to 6b. On the other hand, the rotor 27 is the coil 26.
It is close to a, 26b and the encoder 28. The output of the encoder 28 is the position signal generator 24.
Has been entered in. The output (electrical angle b) of the position signal generator 24 is fed back to the input sides of the deviation compensator 21 and the function generator 23.

Next, the operation will be described. Since this step motor is driven in high resolution microsteps, sufficient linearity must be ensured. Therefore, the harmonic component included in the torque distribution generated by the step motor needs to be reduced as much as possible (the step motor with the reduced harmonic component is referred to as a micro step motor). FIG. 4 shows the torque distribution of the motor. The figure (A) shows the A-phase torque distribution by the coil 26a, and the figure (B) shows the B-phase torque distribution by the coil 26b. The vertical axis represents the generated torque T and the horizontal axis represents the rotation angle. This step motor employs vernier magnetic pole teeth to reduce harmonics in the generated torque distribution.

FIG. 5 is a conceptual diagram showing the torque generating mechanism of the microstep motor. In FIG. 5, the coils 26a and 26b are two-phase coils, and the vector φ _m indicates the magnetic pole vector of the rotor 27 (however, the rotor 27 has a pair of magnetic poles for simplification, Actually use a multi-pole pair of magnetic poles). The encoder 28 has a function of detecting a relative rotation angle of the magnetic pole vector φ _m with respect to the coils 26a and 26b3a. Here, the currents flowing in the A phase and the B phase are currents I _{A and} I _B
If the generated torque distribution does not include harmonics, the A-phase and B-phase torques T _{a and} T _b are represented by the following equations. _{T a = -K T · I A} sin (b) ··· (21) T b = -K T · I B sin (b-π / 2) = K T · I B · cos (b) ··· (22)

Here, according to the electrical angle b of the current position signal and the electrical angle h of the output of the limiter 22, the currents I _A, I _B
Is assumed to have generated a cosine wave current and a sine wave current by the function generator 23 with the operating electrical angle e = b + h as an input. K _T is a coefficient. The currents I _{A and} I _B are respectively expressed by the following equations. I _A = I ₀ · cos (b + h) (23) I _B = I ₀ · sin (b + h) (24) Here, the current I ₀ is a current peak value.

As described above, the combined magnetomotive force φ _i and the virtual coil due to the currents I _{A and} I _B of the coils 26a and 26b are formed at the electric angle e = b + h, as shown in FIG. means. At this time, the total generated torque T is (2
From equations (1) to (24), the following is obtained. T = K _T · I ₀ {−sin (b) · cos (b + h) + cos (b) · sin (b + h)} = K _T · I ₀ · sin (h) ... (25)

Equation (25) is defined by the two-phase coils 26a, 26.
This means that the position of the rotor 27 can be controlled by freely adjusting the phase of the combined magnetomotive force φ _i generated by the current of b. In other words, by controlling the electrical angle h = e−b which is a variable, the generated torque T changes (torque control) and the positioning characteristic is obtained.

FIG. 6 is a characteristic diagram of static torque in the microstep motor. The vertical axis represents the generated torque T, and the horizontal axis represents the operating electrical angle e. (25)
As shown in the equation, the generated torque T is a sine function of the electrical angle h. Therefore, the generated torque T increases as the electrical angle h gradually increases from zero. Then, the point of the electrical angle e = e ₁ in FIG.
When it exceeds = π / 2 [rad], the generated torque T decreases. Further, when the electrical angle h exceeds π [rad] and reaches the point of e = e _{2 in} FIG. 6, the sign of the generated torque is reversed. Then, torque opposite to the command is generated, and the control system becomes unstable. In the worst case, step out. Therefore, it is necessary to limit the absolute value of the electrical angle h so as not to exceed π / 2 [rad], and thus the limiter 22 is indispensable.

If the deviation compensator 21 is subjected to PID control, it is possible to shorten the settling time by electrically providing damping by a differential operation. In addition, the integration operation can eliminate the hysteresis inherent to the stepping motor and the steady deviation due to friction.

Further, the function generator 23 has an operating electric angle e.
Is input and the sine and cosine functions stored in the ROM table are referenced and output. The cogging torque of the micro step motor does not need to be corrected for the contents of the ROM which are sufficiently reduced. However, if the torque distribution contains harmonic components, the contents of the ROM can be changed to make the correction.

Further, the controller will be described.
Conventionally, as a magnetic head positioning control system of a magnetic recording / reproducing apparatus using a sector servo system, there is the following system.

A high-accuracy encoder such as a light transmission type is added to the CPU through an interpolation circuit or an A / D converter.
This is a method of taking in position information in (central processing unit) and performing sample value control (PID control or the like).

Next, FIG. 7 shows an example of the configuration of a conventional typical control system. In FIG. 7, the deviation signal detection unit 71 receives the motor position (true value) x and the eccentricity (true value) r of the medium, and outputs the deviation (true value) detected by the head output from the deviation signal detection unit 71. ) E is input to the deviation compensator 72. The deviation compensator 72 outputs the predicted media eccentricity r ^* . A memory unit 73 is connected to the deviation compensator 72. Then, the input of its eccentricity of predicted media r ^* and the encoder of the detected motor position x ^*, the deviation e ^* is P between the detected motor position eccentric r ^* and the encoder of the predicted media x ^*
It is input to the ID compensator 74. Then, the output of the PID compensator 74 and the output of the position signal detector 77 are input to the pulse motor 76. The motor position (true value) x is output from the pulse motor 76, and the motor position (true value) x is fed back to the deviation compensator 72. Further, the motor position x is detected by the position signal detection unit 77,
It is fed back to the PID compensator 74 and the pulse motor 76. Then, the deviation compensator 72 and the PID compensator 74
And constitute the controller 75.

The operation of the tracking control system in the large-capacity FDD having the above configuration will be described. Motor position (true value) x, media eccentricity r, encoder displacement −
Except for the output gain Ke, the deviation e detected by the head can be detected by the system for each sector (every 1.3 msec). Also, the motor position detected by the encoder x ^*
The eccentricity r ^{* of} the medium predicted to be can be detected by the system side every minor sampling time (every 0.2 msec). The predicted eccentricity r ^{* of} the medium is learned when the power is turned on and is stored in the memory unit 73 in advance.

Now, the influence of the encoder gain variation on the control system will be described. In the minor servo loop, the main component of eccentricity (15H
The open loop gain in the vicinity of Z, 30HZ) is sufficiently secured (30 dB to 20 dB). Therefore, the following relationship holds. x ^* = r ^* = x · Ke (26) e = r−x (27)

Further, assuming that the predicted media eccentricity (learned media eccentricity) r ^* is equal to the actual media eccentricity r, r ^* = r ... (28) From equations (26) to (27), it is found that a steady deviation of e = | (Ke-1) /Ke|.r ^* ... (29) occurs (in the case of only minor loop).

As described above, when the reflection type encoder is used as the position feedback sensor of the FLPM, its gain fluctuates more than that of the transmission type, so that a large steady deviation occurs when following the eccentricity of the medium.

[0021]

In the sectors where the position error signal from the head is not obtained, the accuracy of the magnetic head following the disk eccentricity greatly depends on the encoder accuracy. If an encoder with high accuracy is used, there is a problem that the tracking accuracy deteriorates.

The present invention has been made in view of such circumstances, and an object thereof is to compensate for fluctuations in encoder gain, improve tracking accuracy, and reduce encoder cost.

[0023]

According to the magnetic head positioning control method of the present invention, the motor position (true value) x and the eccentricity (true value) r of the medium are input, and the deviation (true value) detected by the head is calculated for each sector. Value) e, a deviation signal detector 1 that outputs the deviation e detected by the head, and a deviation compensator 2 that outputs the predicted media eccentricity r ^* , and the predicted media eccentricity r ^*. A memory unit 3 for learning and storing in advance, a position signal detector 9 for obtaining a motor position x ^* detected by the encoder, a deviation e detected by the head,
Based on the predicted eccentricity r ^{* of} the medium and the motor position x ^* detected by the encoder, the displacement-output gain Ke of the encoder and a correction coefficient He which is the reciprocal thereof.
An encoder gain compensator 4 for sequentially calculating the deviation, a PID compensator 5 to which a deviation e ^* between the predicted media eccentricity r ^* and the motor position x ^* detected by the encoder is input, and the encoder gain compensator 4 and the output of the position signal detection unit 9 are input, and a multiplier 6 that multiplies the motor position x ^* detected by the encoder by the correction coefficient He
And a pulse motor 8 to which the output of the PID compensator 5 and the output of the multiplier 6 are input and which outputs the motor position (true value) x.

[0024]

In the magnetic head positioning control system of the present invention, the deviation signal detecting section 1 is supplied with the motor position (true value) x and the eccentricity (true value) r of the medium, and the deviation signal detecting section 1 outputs each sector. The deviation (true value) e detected by the head is output. Then, the deviation e detected by the head is input to the deviation compensator 2, and the predicted eccentricity r ^{* of} the medium is output from the deviation compensator 2. Then, the predicted eccentricity r ^{* of} the medium is previously learned and stored in the memory unit 3. On the other hand, the position signal detector 9 obtains the motor position x ^* detected by the encoder. Then, in the encoder gain compensator 4, the deviation e detected by the head,
Based on the predicted eccentricity r ^{* of} the medium and the motor position x ^* detected by the encoder, the displacement-output gain Ke of the encoder and a correction coefficient He which is the reciprocal thereof.
Is calculated sequentially. Then, the deviation e ^* between the predicted eccentricity r ^{* of} the medium and the motor position x ^* detected by the encoder is input to the PID compensator 5. And
The outputs of the encoder gain compensator 4 and the position signal detector 9 are input to the multiplier 6. Then, in the multiplier 6, the motor position x ^* detected by the encoder is multiplied by the correction coefficient He. And the pulse motor 8
The output of the PID compensator 5 and the output of the multiplier 6 are input to. Then, the motor position (true value) x is output from the pulse motor 8. By the above, the fluctuation of the encoder gain is compensated and the tracking accuracy is improved.

[0025]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A large capacity FDD using a sector servo system will be described by taking as an example a case where a linear pulse motor is used as a magnetic head actuator.

In addition to this, the linear encoder may have a scale externally attached to the mover.

FIG. 1 shows a configuration example of the present invention. The present invention
Is the magnetic field of a magnetic recording / reproducing device using the sector servo system.
Qi head positioning control method. In Figure 1, the deviation
The signal detection unit 1 has a motor position (true value) x and a media
The eccentricity (true value) r is input, and the deviation signal detection unit 1
The deviation (true value) e detected by the head output from the
It is input to the compensator 2. From the deviation compensator 2
Eccentricity r ^* Is output. Also, the above bias
The memory unit 3 is connected to the difference compensator 2. That
The deviation e detected by the head and the predicted media
Eccentricity r^* Is input to the encoder gain compensator 4. Well
Predicted media eccentricity r^* And encoder detection
Motor position x^* Enter the predicted media
A eccentricity r^*And the motor position detected by the encoder x^* When
Deviation e^* Are input to the PID compensator 5. And
The output of the PID compensator 5 and the output of the multiplier 6 are pulsed.
It is input to the motor 8.

The output from the encoder gain compensator 4 is input to the multiplier 6. The deviation compensator 2, the memory unit 3, the PID compensator 5, the encoder gain compensator 4, and the multiplier 6 form a controller 7. Also,
The motor position (true value) x is output from the pulse motor 8, and the motor position x is fed back to the deviation compensator 2. The motor position x is determined by the position signal detection unit 9
To the multiplier 6, and the output of the position signal detector 9 is input to the multiplier 6.

Next, the operation will be described. The deviation signal detector 1 obtains the deviation e from the head for each sector. The deviation e is input to the deviation compensator 2. Then, in the memory unit 3, the amount of eccentricity is learned and stored in advance. Further, the position signal detector 9 obtains the motor position x ^* detected by the encoder. Then, in the encoder gain compensator 4, based on the deviation e detected by the head, the predicted eccentricity r ^{* of} the medium stored in the memory unit 3, and the motor position x ^* detected by the encoder, Displacement-output gain Ke and its reciprocal correction coefficient He
Is calculated sequentially. Then, in the multiplier 6, the motor position x ^* detected by the encoder is multiplied by the correction coefficient He. The correction coefficient He cancels the influence of the gain variation of the encoder.

Next, the relationship between the signals will be described. Since the predicted eccentricity r ^{* of} the medium (eccentricity information stored in the memory in advance) and the motor position x ^* detected by the encoder (positional information obtained from the encoder) act so as to be equal, Relationship is established. x ^* = r ^* = Ke · x (1) e = r−x (2)

Further, by frequently performing the storage operation of the eccentricity amount in R / W free time, etc., the predicted media eccentricity r ^* and the media eccentricity r (actual eccentricity amount) can be regarded as equal. r ^* = r ... (3) As described above, the displacement-output gain Ke of the encoder can be obtained from the equations (1), (2), and (3) as follows. Ke = x ^* / x = x ^* / (r ^* -e) (4)

Therefore, encoder displacement-output gain K
The correction coefficient He for e can be obtained by multiplying the correction coefficient He ′ obtained in the preceding sector by the correction coefficient He (= Ke ⁻¹ ) obtained this time. Therefore, He = He ′ · He = He ′ · (r ^* −e) / x ^* ... (5) However, the initial value of He is 1 here. The encoder gain compensator 4 operates every time the deviation e detected by the head is read, and rewrites the correction coefficient He.

Next, the encoder gain fluctuation allowable value when this control method is applied will be described. In the first-order mode, the frequency is 15 HZ, the eccentricity amount is ± 16 μm, and the maximum displacement amount in one sector is 2.01 μm. In the secondary mode, the frequency is 30 HZ, the eccentricity amount is ± 10 μm, and the maximum displacement amount in one sector is 2.51 μm. Also, the media rotation speed is 900 rpm and the number of sectors is 50 sectors / r.
rotation, head sampling f is 750HZ. Therefore, the maximum eccentricity per sector in each mode is 16 · sin (2π · 15) in the primary mode.
/750·1/2)·2=2.01 μm. Also,
Similarly, in the secondary mode, 10 · sin (2π · 30/7)
50 · 1/2) · 2 = 2.51 μm. Also, Ke
From the error distribution such as the eccentricity compression ratio and the vibration resistance when = 1, assuming that the errors generated by the displacement of the encoder in each mode-the output gain Ke change are errors e ₁ and e ₂ , e ₁ ² + e ₂ ² = 0.89 μm (6)

Here, the gain variation ratio Z is defined below. FIG. 2 is a diagram showing the relationship between the encoder output value and the displacement. In the same figure, the inclination (gain) of the encoder output for an arbitrary 3 μm displacement is k _n , and the inclination for an adjacent 3 μm displacement is k _{n + 1} . Then, the gain variation ratio Z is defined as Z = k _{n + 1} / k _n .

Therefore, the gain variation ratio Z in each mode is
When the same encoder signal is used, since they are the same, e ₁ /2.01=e ₂ /2.51=Z (7) Then, from equations (6) and (7), e ₁ = 0.5
90 μm, e ₂ = 0.736 μm, and Z = ± 29.4%. Therefore, the specification is 0.1 μm unit,
e ₁ = 0.6 μm, e ₂ = 0.7 μm, and Z = ± 28%.

As described above, the off-track amount (deviation e detected by the head) obtained from the head for each sector, a prediction signal learned at the time of initialization (predicted media eccentricity r ^* ), and further an encoder signal. The displacement-output gain Ke of the encoder can be estimated based on (the motor position x ^* detected by the encoder). By multiplying the encoder signal by the reciprocal thereof, the gain variation of the encoder is compensated for each sector. Therefore, the influence of the gain variation is not accumulated over the next sector, and the variation between the sectors may be suppressed. Then, the deterioration of the tracking accuracy due to the gain variation of the encoder is compensated, and the tracking accuracy is improved. Also, the cost of the encoder is reduced.

By applying the method of the present invention, the gain variation ratio of the encoder may be ± 28% (every 0.3 μm pitch), and the current accuracy satisfies the specifications. The current ability is ± 10%.

[0038]

As described above, according to the magnetic head positioning control method of the present invention, the encoder displacement compensator sequentially calculates the encoder displacement-output gain Ke and the reciprocal of the correction coefficient He. Then, in the multiplier, the motor position x ^* detected by the encoder is multiplied by the correction coefficient He, so that the fluctuation of the encoder gain is compensated even when a low-precision encoder with large gain fluctuation is used at low cost. can do. Then, the tracking accuracy can be improved and the cost of the encoder can be reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of a magnetic head positioning control system of the present invention.

FIG. 2 is a diagram showing a relationship between an encoder output value and a displacement.

FIG. 3 is a block diagram showing a configuration of an example of a conventional closed loop servo system of a step motor.

FIG. 4 is a diagram showing a distribution of torque generated by the motor. FIG. 4A shows an A-phase torque distribution by the coil 26a, and FIG. 4B shows a B-phase torque distribution by the coil 26b.

FIG. 5 is a conceptual diagram showing a torque generation mechanism of a micro step motor.

FIG. 6 is a characteristic diagram showing static torque in a microstep motor.

FIG. 7 is a block diagram showing a configuration of an example of a conventional magnetic head positioning control system.

[Explanation of symbols]

1 deviation signal detection unit 2 deviation compensator 3 memory unit 4 encoder gain compensator 5 PID compensator 6 multiplier 7 controller 8 pulse motor 9 position signal detection unit r media eccentricity (true value) e head detected deviation (true) Value) r ^* Predicted media eccentricity x ^* Motor position detected by encoder e ^* Deviation between predicted media eccentricity r ^* and motor position x ^* detected by encoder x Motor position (true value) He correction coefficient Ke encoder displacement-output gain

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Internal reference number FI technical display location G11B 21/10 Q 8425-5D

Claims (1)

[Claims] 1. A magnetic recording / reproducing apparatus using a sector servo system for positioning the rotor by detecting the position of the rotor of the pulse motor with an encoder and controlling the magnetomotive force phase of the stator with respect to the rotor. In the magnetic head positioning control method, the motor position (true value) x and media eccentricity (true value) r
And a deviation signal detection unit that outputs the deviation (true value) e detected by the head for each sector, and a deviation that outputs the deviation e detected by the head and outputs the predicted eccentricity r ^{* of} the medium. A compensator, a memory unit for learning and storing the predicted eccentricity r ^{* of} the medium in advance, a position signal detecting unit for obtaining the motor position x ^* detected by the encoder, a deviation e detected by the head, Based on the predicted eccentricity r ^{* of} the medium and the motor position x ^* detected by the encoder, the encoder displacement-output gain Ke
And an encoder gain compensator that sequentially calculates a correction coefficient He that is the reciprocal thereof, and a deviation e ^* between the predicted media eccentricity r ^* and the motor position x ^* detected by the encoder is input PI
The D compensator, the encoder gain compensator and the output of the position signal detector are input, the multiplier that multiplies the motor position x ^* detected by the encoder by the correction coefficient He, the output of the PID compensator and the A magnetic head positioning control method, comprising: a pulse motor which receives the output of the multiplier and outputs the motor position (true value) x.