JPH10149643A - Magnetic head positioning controller - Google Patents

Abstract

PROBLEM TO BE SOLVED: To shorten an accessing time and to improve head positioning accuracy without taking any special positioning/seeking conditions for calibration. SOLUTION: The acceleration constant compensator 24 of a positioning controller 1 receives an A/D converted position signal 2, an A/D converted current detecting signal 3, an estimated speed signal 16 outputted from a condition estimator 17 and an external force compensating signal 15 outputted from an external force compensator 23, placing these signals in an acceleration constant measuring expression and outputs an acceleration constant corrected value 25 to the condition estimator 17, a speed controller 18 and a phase controller 19.

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 controller (hereinafter, simply referred to as "positioning controller") used in a magnetic disk drive.

[0002]

2. Description of the Related Art A positioning controller of this kind performs a high-speed, high-accuracy, stable positioning control of a magnetic head as disclosed in, for example, JP-A-6-215508 and JP-A-6-243611. It is intended for.

FIG. 8 is a functional block diagram showing a conventional positioning controller. Hereinafter, description will be made based on this drawing.

A magnetic disk drive includes a positioning controller 1a.
In addition, a D / A converter 9 for converting an output signal of the positioning controller 1a from a digital signal to an analog signal and outputting the converted signal as a VCM drive signal 4, a servo amplifier 10 for amplifying the VCM drive signal 4, An actuator 11 for positioning a magnetic head 13 by a servo output signal,
A position error signal generating circuit 6 for demodulating the position information 5 on the magnetic disk 12 read by the magnetic head 13 as a position signal 2, an A / D converter 7 for converting the position signal 2 from an analog signal to a digital signal, An A / D converter 8 for converting the current detection signal 3 from the amplifier 10 from an analog signal to a digital signal is provided.

The positioning controller 1a includes a state estimator 17 for estimating control information, a phase compensator 19 for precise positioning, a speed controller 18 for speed following control, a control switching unit 14, a settling unit The variation of the acceleration constant (K _F / m [m / s ² / A]) with respect to the actuator 11 including the end determination unit 20, the external force compensator 23, and the voice coil motor (hereinafter, referred to as “VCM”). And an acceleration constant compensator 24a for correction.

Next, the magnetic head positioning operation will be described.

First, the position information 5 on the magnetic disk 12 read by the magnetic head 13 is stored in a position error signal generation circuit 6.
Is converted into a position signal 2 via The A / D converter 7 discretizes the demodulated position signal 2, and the A / D converter 8 discretizes the current detection signal 3 from the servo amplifier 10. The positioning controller 1a is a digital signal processor (not shown; one of the microprocessors; hereinafter, referred to as “DSP”).
That. ) Processed by the firmware and output. This output signal is held by the D / A converter 9 and then output to the servo amplifier 10 as the VCM drive signal 4. The current detection signal 3 output from the servo amplifier 10 drives the actuator 11 to move the magnetic head 13 to a target cylinder (track) on the magnetic disk 12. Further, the state estimator 1 constituting the positioning controller 1a
7. The phase compensator 19, the speed controller 18, the control switch 14 for switching between the speed following control and the fine positioning control, the settling end determiner 20, the external force compensator 23, and the acceleration constant compensator 24a are composed of DSP firmware (program ) Is realized inside.

This DSP starts a seek when it receives a seek command from a higher-level device (not shown). More specifically, the seek operation is performed by using the speed from the current position to the target position of the magnetic head 13 as seek length information, by a speed following control by a speed controller 18 to a target speed corresponding to a seek length, and by a phase compensator 19 by a target. It performs precision positioning control near the cylinder. The speed following control and the precise positioning control are switched by the control switch 14 when the control cylinder 14 reaches the vicinity of the target cylinder. The end of the seek is determined by the settling end determiner 20 comparing the target position with the current position signal, and is output to the host device.

A state estimator 17 receives the position signal 2 discretized by the A / D converter 7 and the current detection signal 3 discretized by the A / D converter 8, and outputs an estimated speed signal 16
Is output. The estimated speed signal 16 is a signal used for the speed following control and the precise positioning control, and is a signal for estimating the moving speed of the magnetic head 13 which is a state estimation amount that cannot be directly observed. This estimation is performed based on the theory of the state estimator of the control theory (observer of Re-Emberger). The outline will be described. First, a digital model to be controlled is created. This model inputs a control current and has speed and position as internal variables. Then, variables when the same input as that of the actual control target is added to the model are simulated in real time. However, since the error diverges as it is due to the difference between the model and the actual control target, the actual measurement position and the estimated position are compared and the error is fed back to an internal variable so that the error converges.

The phase compensator 19 inputs the discretized position signal and the estimated speed signal 16 and outputs a control signal so that the control system becomes stable during the precise positioning control. The speed controller 18 obtains seek length information from the target cylinder input from the host circuit, and outputs a control signal proportional to the difference between the target speed from the seek length information and the estimated speed signal 16. The external force compensator 23 has a different V depending on the cylinder position.
Holding the external force applied to the CM as a table (not shown)
A control signal is output so as to cancel external force.

The acceleration constant compensator 24a has a parameter table (not shown) obtained from the VCM acceleration constant (K _F / m) that varies depending on the cylinder position, and calculates the difference from the acceleration constant used for design during positioning and seek. Is output to the state estimator 17, the speed controller 18, and the phase compensator 19 as a correction gain. In addition, the acceleration constant is measured by an instruction of a higher-level circuit (calibration instruction 26), and the parameter table is adjusted based on the measurement result, thereby reducing the speed following error at the time of high speed following control and positioning at the time of precise positioning control. Reduce the error. Specifically, as shown in Japanese Patent Application Laid-Open No. 6-259900, a minute sine wave is added to a VCM drive signal, and the acceleration constant is measured from the amount of change of the position signal with respect to the VCM drive signal. And the variation of the acceleration constant due to aging.

[0012]

In the prior art, the acceleration constant compensator had to take a special positioning state for calibration when measuring the acceleration constant of the VCM. Specifically, a minute sine wave is added to the VCM drive signal to measure the acceleration constant from the amount of change in the position signal with respect to the minute sine wave, or to perform a trial seek for measuring the acceleration of the VCM. The value of the acceleration constant fluctuates due to temperature changes and aging during operation of the VCM,
Remeasurement must be performed over time.

The problem here is that it is necessary to take a special positioning and seek state for calibration.
The response from the host device becomes slow. In the positioning controller, it is most important to improve the head positioning accuracy and shorten the access time.

The positioning controller controls the magnetic head by switching between high-speed follow-up control for shortening the head access time and precision positioning control for improving the head positioning accuracy. Further, in these controls, what is directly observed are the position information and the drive current, and the head moving speed is estimated by the state measuring device.

[0015]

SUMMARY OF THE INVENTION An object of the present invention is to provide a positioning controller capable of realizing both shortening of access time and improvement of head positioning accuracy without taking a special positioning / seek state for calibration. .

[0016]

In order to achieve the above object, it is desirable to measure an acceleration constant during a normal positioning / seek operation. The positioning controller of the present invention performs an ordinary seek operation without taking a special positioning / seek state for calibration, thereby obtaining an acceleration constant (K _F / m [m / s) of an actuator (for example, VCM).
² / A]). More specifically, the A / D-converted position signal of the magnetic head, the A / D-converted current detection signal, the external force compensation signal obtained by the external force compensator, and the estimation obtained by the state estimator. An acceleration constant compensator for calculating an acceleration constant of the VCM based on the speed signal;

According to the present invention, the access time can be reduced and the head positioning can be reduced by outputting a VCM driving signal which is a head positioning control signal by a state estimator, a speed controller and a phase compensator using the measured VCM acceleration constant. It is possible to simultaneously achieve both improvement in accuracy. Further, since there is no need to set a special positioning state for measuring the acceleration constant, it is possible to measure the acceleration constant without delaying the response as viewed from the host device.

[0018]

FIG. 1 is a functional block diagram showing an embodiment of a positioning controller according to the present invention. Hereinafter, description will be made based on this drawing. However, the same parts as those in FIG.

The positioning controller 1 of this embodiment is the same as the acceleration constant compensator 24a of the conventional positioning controller 1a (FIG. 8).
An acceleration constant compensator 24 is provided instead of (FIG. 8).

The position information 5 on the magnetic disk 12 read by the magnetic head 13 is converted into a position signal 2 via a position error signal generation circuit 6. The A / D converter 7 discretizes the demodulated position signal 2, and the A / D converter 8 discretizes the current detection signal 3 from the servo amplifier 10. The positioning controller 1 processes and outputs with the DSP firmware. The output signal is held by the D / A converter 9 and then output to the servo amplifier 10 as the VCM drive signal 4. Then, the current detection signal 3 output from the servo amplifier 10
Drives the actuator 11 constituted by the VCM to move the magnetic head 13 to a target cylinder (track) on the magnetic disk 12. The DSP includes a state estimator 17 and a phase compensator 1 that constitute the positioning controller 1.
9. The operations of the speed controller 18, the control switch 14 for switching between the speed following control and the precision positioning control, the external force compensator 23, and the acceleration constant compensator 24 are controlled by firmware.

The acceleration constant compensator 24 of the positioning controller 1
Are the A / D converted position signal 2, the A / D converted current detection signal 3, the estimated speed signal 16 output from the state estimator 17, and the external force compensation signal output from the external force compensator 23. 15 are input into the acceleration constant measurement formula, and the acceleration constant correction value 25 is output to the state estimator 17, the speed control 18 and the phase controller 19.

FIG. 2 is a flowchart showing a first example of a procedure for measuring the acceleration constant of the DSP in the positioning controller of the present embodiment. Hereinafter, description will be given based on FIG. 1 and FIG.

When the DSP receives a seek command from the host device, it starts seeking (step A1). First, to record the position x _k0 of the seek start in memory (not shown) (step A2), is cleared to 0 to S _V and S _u performing summation (step A3). When the seek operation is started, an estimated speed signal 16 which is an output signal of the state estimator 17, an output signal of the speed controller 18, and an output signal of the phase compensator 19 are respectively calculated based on the acceleration constant of the VCM which has been obtained so far. Then, the external force compensating signal 15 which is the output signal of the external force compensator 23 is combined with the head positioning control signal VC
An M drive signal 4 is output. Thus, a seek is performed, and the magnetic head 13 is moved to the target position received from the host device (step A4). Further, while calculating the VCM driving signal 4, the estimated speed signal 16 during the seek is added to S _v, the difference between the current detection signal 3 and the external force compensation signal 15 S _u
(Step A5). When the seek ends, the position x _k1 at the end of the seek is recorded in the memory (step A).
6, A7). Further, the acceleration constant of the VCM is calculated by the following equation (step A8), and the acceleration constant correction value 2 measured by the state estimator 17, the speed controller 18 and the phase compensator 19 is calculated.
5 is output (step A9).

K _F / m = {(x _k1 −x _k0 ) −T × S _v } / {T ² × S _u / 2} (1)

Next, equation (1) will be described. The state equation of the discrete system of the magnetic head of the magnetic disk drive is as follows: x is the position signal, v is the moving speed of the magnetic head, i is the current detection signal, n is the external force correction signal, T is the sampling time, and k is the sampling time. It is expressed by the following equation.

[0026]

(Equation 1)

When the state equation of equation (2) is modified, the following equation is obtained. Here, k0 is the sampling time at the start of the seek, and k1 is the sampling time at the end of the seek.

[0028]

(Equation 2)

[0029]

[0030]

(Equation 3)

Then, the equation for measuring the acceleration constant of the acceleration constant compensator 24 is as shown in equation (1).

Therefore, by using equation (1),
It is possible to measure the acceleration constant during a normal seek operation.

FIGS. 3 and 4 are flowcharts showing a second example of the procedure for measuring the acceleration constant of the DSP in the positioning controller of the present embodiment. Hereinafter, a description will be given based on FIGS. 1, 3 and 4.

When the DSP receives a seek command from the host device, the DSP starts seeking (step B1). First, the position x _k00 at the start of the seek is recorded in a memory (not shown) (step B2), and S _v0 , S _u0 , S _v1, and S _u1 for performing the sum calculation are cleared to 0 (step B3). S _v0 and S _u0 are for calculation in the first half of the seek, and S _v1 and S _u1 are for calculation in the second half of the seek. When the seek operation is started, the estimated speed signal 16 which is the output signal of the state estimator 17, the output signal of the speed controller 18, and the output signal of the phase compensator 19 are calculated by the acceleration constant of the VCM which has been obtained so far. A VCM drive signal 4 as a head positioning control signal is output in accordance with the external force compensation signal 15 output from the external force compensator 23. Thus, a seek is performed, and the magnetic head 13 is moved to the target position received from the host device (step B).
4). Further, while calculating the VCM drive signal, the estimated speed signal 16 during seek is added to S _v0 , and the current detection signal 3
_Is added to S _u0 (step B).
5).

[0035] Then, the recording position x _{k0 1} during sampling the magnetic head 13 has passed the intermediate position of the seek (= x _{k1 0)} in the memory (step B6, B7), the formula (1) by the seek first half of the Calculate a first acceleration constant (step B
8). Further, while calculating the VCM drive signal, the estimated speed signal 16 during seeking is added to S _v1 , and the difference between the current detection signal 3 and the external force compensation signal 15 is added to S _u1 (step B9). When the seek ends (step B10), the position x _k11 at the end of the seek is recorded in the memory (step B1).
1) A second acceleration constant in the latter half of the seek is calculated by equation (1) (step B12).

Subsequently, the average value of the first acceleration constant value in the first half of the seek and the second acceleration constant value in the second half of the seek is calculated as the VCM acceleration constant in the seek range (step B13), and the state estimator 17 is calculated. Then, the measured acceleration constant correction value 25 is output to the speed controller 18 and the phase compensator 19 (step B14).

In this example, the calculation is performed by dividing the acceleration constant into two at the middle position of the seek length. However, the position of the head divided into the first half and the second half of the seek need not be the middle position of the seek. Further, the acceleration constant of each seek portion may be calculated more finely by dividing the seek portion into a plurality of seek portions, and the acceleration constant of the seek range may be obtained from the average value.

Next, the parameter table of the acceleration constant in the DSP is updated from the measured acceleration constant (K _F / m), and the state estimator 17, the speed controller 18, and the phase compensator 1 are updated.
Next, a method of outputting the data to 9 will be described.

The VCM constituting the actuator of the magnetic disk drive has a different acceleration constant depending on the position (cylinder). For this reason, it is assumed that the DSP divides all cylinders into blocks each having a fixed number of cylinders, has a parameter table of acceleration constants for each block, and that the initial value is measured in advance for each block by a conventional method. Equation (1)
As can be seen from the right side of the equation, in the acceleration constant measuring method of the acceleration constant compensator of the present invention, since the measurement result of the acceleration constant is a function of the seek start position x _k0 and the end position x _k1 , the measurement result is the acceleration constant. Must be output to a parameter table consisting of multiple blocks.

FIG. 5 is an explanatory diagram when the DSP uses the measurement result of the acceleration constant in the positioning controller of the present embodiment. FIGS. 6 and 7 are flowcharts showing a processing procedure in which the DSP uses the measurement result of the acceleration constant in the positioning controller of the present embodiment. Hereinafter, a description will be given based on FIGS. 5, 6, and 7.

For each block, the DSP calculates Y _i (i = _i ) for the central cylinder x _i (i = 0, 1,...) Of the block.
., 0, 1,...) Is assumed to have a parameter table of acceleration constants in the memory. For simplicity, the seek start position (cylinder) x _k0 is smaller than the seek end position (cylinder) x _k1 , that is, x _k0 ≦ x _k1 .

When the DSP completes the measurement of the acceleration constant according to the equation (1), the DSP determines in which block the position x _k0 at the start of the seek is located, ie, the cylinders x _i-1 and x _i (where i
≧ 1) (Steps C0 to C)
2). Next, from the positional relationship between x _i-1 and x _k0 , x _i , an acceleration constant Y _k0 of x _k0 with reference to the parameter table is obtained by a proportional calculation (step C3). Similarly, in which block the position x _k1 at the end of the seek is located, that is, x
determine whether the _j-1 and x _j (except j ≧ i) (step C4, C5, C8), determine the acceleration constant Y _k1 of x _k1 when referring to the parameter table from its positional relationship with proportional calculation (Step C9). At this time, Y _k0 , Y _i ,
The cylinder integral value of the acceleration constant is obtained by trapezoidal approximation from Y _{i + 1} ,..., Y _j , Y _k1 , and the obtained cylinder integral value (approximate value)
Is divided by the seek distance to obtain the integral average (step C
6, C7, C10). Subsequently, the difference between the acceleration constant measurements Y _m measured integrated average and the parameter table of the acceleration constant between the seek start cylinder x _k0 and seek ending cylinder _{x k1 Y p (p = i} -1, i, i + 1, …, J, j +
1) (steps C11 to C14).

As a result, the acceleration constant measurement result can be expanded in the acceleration constant parameter table, and the difference between the acceleration constant used in the design at the time of positioning and seeking is used as a correction gain as the state estimator 17, speed controller 18, Output to the phase compensator 19. Therefore, by updating the parameter table of the acceleration constant for each seek, it is possible to reduce the acceleration tracking error in the high-speed tracking control and the positioning error in the precision positioning control. In FIG. 5, the difference between the integrated average of the acceleration constants from x _k0 to x _k1 before measurement and the measurement result of the acceleration constant is equal to the difference between the acceleration constant curves before and after the measurement.

According to the above method, there is no need to set a special positioning state for measuring the acceleration constant.
Is a state estimator 17, a speed controller 18, a phase compensator 19
In addition, the external force compensator 23 can realize high-speed, high-accuracy, stable positioning control using the measured acceleration constant.

Although the acceleration constant has been described as a unit of [m / s ² / A], the unit can be changed to a unit convenient for calculation and does not limit the applicable range of the present invention. For example, [rad / s ² / A] or [cyl / T ² /
A] (cyl is a cylinder width, and T is a sampling cycle).

[0046]

According to the positioning controller of the present invention, the acceleration constant can be measured during the normal seek operation,
Since the magnetic head can be accurately positioned without delaying the response as viewed from the host device, the access time can be reduced and the positioning accuracy can be improved.

[Brief description of the drawings]

FIG. 1 is a functional block diagram showing an embodiment of a positioning controller of the present invention.

FIG. 2 is a flowchart showing a first example of a procedure for measuring an acceleration constant in the positioning controller of the present invention.

FIG. 3 is a flowchart showing a second example of a procedure for measuring an acceleration constant in the positioning controller of the present invention.

FIG. 4 is a flowchart showing a second example of a procedure for measuring an acceleration constant in the positioning controller of the present invention.

FIG. 5 is an explanatory diagram in a case where a measurement result of an acceleration constant is used in the positioning controller of the present invention.

FIG. 6 is a flowchart when a measurement result of an acceleration constant is used in the positioning controller of the present invention.

FIG. 7 is a flowchart when a measurement result of an acceleration constant is used in the positioning controller of the present invention.

FIG. 8 is a functional block diagram showing a conventional positioning controller.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Positioning controller 2 Position signal 3 Current detection signal 4 VCM drive signal 5 Position information 6 Position error signal generation circuit 7, 8 A / D converter 9 D / A converter 10 Servo amplifier 11 Actuator 12 Magnetic disk 13 Magnetic head 14 Control switch 15 External force compensation signal 16 Estimated speed signal 17 State estimator 18 Speed controller 19 Phase compensator 20 Settling end determiner 21 Settling end signal 22 Target position 23 External force compensator 24 Acceleration constant compensator 25 Acceleration constant correction value

Claims (7)

[Claims] 1. A state estimator for outputting an estimated speed signal corresponding to a moving speed of a magnetic head, a speed controller for outputting a speed following control signal for causing the magnetic head to follow a target speed, and a magnetic head positioning device An external force compensator for outputting an external force compensation signal for canceling an external force applied to the actuator, a phase compensator for outputting a precise positioning control signal for precisely positioning the magnetic head near a target cylinder, and an acceleration constant of the actuator And an acceleration constant compensator that adjusts the gain of the state estimator, the speed controller, and the phase compensator based on the measurement result, based on the estimated speed signal, the speed following control signal, and the precise positioning control signal. In a magnetic head positioning controller that outputs a drive signal to the actuator, the acceleration constant compensator includes: Position signal in the click, the driving current detection signal of said actuator, based on said estimated speed signal and the external force compensating signal to measure said acceleration constant, the magnetic head positioning control device, characterized in that. 2. The acceleration constant compensator is obtained by subtracting the external force compensation current from the seek end position x _k1 , the seek start position x _k0 , the sampling period T, the total estimated speed S _v , and the drive current of the actuator. Assuming that the sum of the objects is S _u and the acceleration constant is K _F / m, K _F / m = {(x _k1 −x _k0 ) −T × S _v } / {T ² × S _u / 2} (1) 2. The magnetic head positioning controller according to claim 1, wherein the average value of the acceleration constant in the seek range is calculated by the following. Wherein said acceleration constant compensator seek end position x _k11, the seek intermediate position x _k10 = x
_k01 , seek start position x _k00 , sampling period T
And then, in a range of x _k01 ~x _k00, the sum of the estimated speed S _v0, S _u0 the sum total obtained by subtracting the external force compensating current from the drive current of the actuator, the acceleration constant is _{K F0 / m, K F0 /} m = {(X _k01 −x _k00 ) −T × S _v0 } / {T ²
× S _u0 / 2} by calculating the average value of the acceleration constant in the first half of the seek range, the range of x _k11 ~x _k10, the sum of the estimated speed S _v1, but minus the external force compensating current from the drive current of the actuator summing S _u1, the acceleration constant is _{K F1 / m, K F1 /} m = {(x k11 -x k10) -T × S v1} / {T 2
The average value of the acceleration constant in the latter half of the seek range is calculated from × S _u1 / 2｝. Further, assuming that the acceleration constant is K _F / m, K _F / m = (K _F1 / m + K _F0 / m) / 2 2. The magnetic head positioning controller according to claim 1, wherein an average value of the acceleration constant in the entire seek range is calculated. 4. The acceleration constant compensator divides a seek range into a plurality of ranges, calculates the acceleration constant according to the equation (1) for each of the divided seek ranges, and seeks an average value of these acceleration constants. The magnetic head positioning controller according to claim 2, wherein the magnetic head positioning controller calculates the average value of the acceleration constant over the entire range. 5. A parameter table in which all cylinders are divided into a plurality of blocks, and an acceleration constant is stored in advance for each block. The acceleration constant compensator includes: an acceleration constant measured during a seek operation; The magnetic head positioning controller according to claim 2, wherein a difference from the acceleration constant stored in the magnetic head is used as a correction amount in a seek range used for measurement. 6. A parameter table in which all cylinders are divided into a plurality of blocks, and an acceleration constant is stored in advance for each block. The acceleration constant compensator includes: an acceleration constant measured during a seek operation; 4. The magnetic head positioning controller according to claim 3, wherein a difference from the acceleration constant stored in the controller is used as a correction amount in a seek range used for measurement. 7. A parameter table in which all cylinders are divided into a plurality of blocks, and an acceleration constant is stored in advance for each block. The acceleration constant compensator includes: an acceleration constant measured during a seek operation; 5. The magnetic head positioning controller according to claim 4, wherein a difference from the acceleration constant stored in the controller is used as a correction amount in a seek range used for measurement.