JPH11120722A - Magnetic disk device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress the influence of variance in an actuator state amount at the time of controller switching in a head position setting operation by using a position controller included in a primary phase advance lag compensator to calculate the initial value of the phase advance lag compensator at the time of switching from a speed controller to the position controller based on the direct transformation of the state estimated amount of an actuator. SOLUTION: For moving a head 4, a control system switching device 8 makes switching between a seek controller 9 and a follow controller 10 and then performs positioning in a target location. The control system switching device 8 performs the switching by detecting a position difference smaller than a specified between the current position and the target position of the head 4. The seek controller 9 is a speed controller for moving the head 4 near the target position, and performs control such that the speed of the head 4 follows a speed profile. The follow controller 10 is a position controller for making the head 4 to follow the target position, and this is composed of a phase advance lag compensator or PID controller.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a magnetic disk device having a feature in head positioning, and more particularly, to a magnetic disk device having a feature in setting a state quantity initial value of a controller when switching the controller in a head access operation. Related to the device.

[0002]

2. Description of the Related Art In a magnetic disk drive, a head is moved to a predetermined track to record information at a predetermined position on a recording surface of a magnetic disk or to read information already recorded. Control is being performed. This positioning control is generally performed by switching a plurality of control systems according to a position deviation which is a difference between the position of the head and the target position. That is, when the position deviation is large, for example, when the position deviation is one track or more, the speed control is performed so that the head speed follows the target speed according to the position deviation, and when the head approaches the target position. Positioning control is performed by position control for controlling the head position so that the position deviation becomes zero. This is an effective method for moving and following the head to the target position with high speed and high accuracy. As described above, when the controller is switched from the speed control to the position control, particularly when the initial setting of the position controller when switching to the position control is not appropriate, the head becomes vibrating due to a transient response and the target position is changed. The time required for positioning to the position becomes longer.

When a PID controller that feeds back the integrated value of the head position deviation, head speed, and head position deviation is used as the position controller, the head position deviation and head speed at the time of switching can be directly transferred. Thus, the feedback gain may be determined so as to optimize the response characteristics of the PID controller. However, when a phase lead / lag compensator is used as a position controller to reduce the amount of calculation, it is necessary to set an appropriate value as an initial setting value of the state quantity.

Conventionally, several methods have been known to solve such a problem. For example, JP-A-4-335
No. 272 discloses an internal part of a phase compensator based on a coefficient calculated in advance corresponding to a plurality of regions on a phase plane classified according to a head position deviation and a speed at the time of switching, and a head position deviation and a speed. Calculating the initial value of the state. The coefficients used here are nTs (n is an appropriately determined number of samples, Ts is a sample number) from the time of switching for the head position deviation, head speed, internal variables of the phase compensator, and the appropriately determined weighting coefficient matrix. Time) It is determined so as to minimize the quadratic evaluation function until after the time. That is, the initial value of the internal state of the phase compensator such that the integrated value of the head position deviation and the head speed until the head settles is minimized, and a coefficient obtained by the head position deviation and the head speed at the time of switching. Is used.

[0005]

In the above prior art, since the transient response characteristic changes depending on the weight coefficient matrix in addition to the response of the phase compensator, the trade-off between the settling time and the smooth response is optimized. It was necessary to search the weight coefficient matrix.

[0006]

A magnetic disk drive according to the present invention comprises a head positioning mechanism having an actuator for moving a head for recording and reproducing data on and from a magnetic disk, and a difference between a current position and a target position of the head. A position deviation detector that detects a position deviation through the head, a head state estimator that estimates a head estimated speed and a head estimated position based on the position deviation, and a drive signal to the actuator based on the position deviation. A speed controller that moves the head to a position near the target position by outputting the signal; and a drive signal that is output to the actuator based on the position deviation to cause the head to follow the target position, and a first-order phase advance / delay. A position controller comprising a compensator, the speed controller and the position controller based on the position deviation; A control system switch to be operated by switching, and when switching from the speed controller to the position controller, the state quantity of the position controller based on the drive signal, the position deviation, the head estimated speed and the head estimated position. And an initial value setting device for setting.

In the present invention, a position controller using a first-order phase lead / lag compensator is used, and the initial value of the phase lead / lag compensator at the time of switching from the speed controller to the position controller is set to PID.
Using the characteristics of the compensator, the estimated state of the actuator is directly converted and obtained. This suppresses the influence of the variation in the state quantity of the actuator when the controller is switched in the head position setting operation.

[0008]

1 and 2 show an embodiment of a magnetic disk drive according to the present invention. FIG. 1 is a block diagram showing an overall configuration, and FIG. 2 is a schematic side view showing a head positioning mechanism. . Hereinafter, description will be made based on these drawings.

The magnetic disk drive of this embodiment has a head positioning mechanism (FIG. 2) having a VCM (voice coil motor) 1 as an actuator for moving a head 4 for recording and reproducing data on and from a magnetic disk 7, A position deviation detector 13 for detecting, via the head 4, a position deviation which is a difference between the current position and the target position of the head 4, a head state estimator 12 for estimating a head estimation speed and a head estimation position based on the position deviation. And a seek controller 9 (speed controller) for moving the head 4 close to the target position by outputting a drive signal to the VCM 1 based on the position deviation, and outputting a drive signal to the VCM 1 based on the position deviation. And a follow controller 10 (position controller) comprising a first-order phase lead / lag compensator, Seek controller 9 and follow-controller 10 control system is operated by switching between switcher 8 based on
And an initial value setting unit 13 for setting a state quantity of the follow controller 10 based on a drive signal, a position deviation, a head estimated speed, and a head estimated position when switching from the seek controller 9 to the follow controller 10. It is.

One or a plurality of magnetic disks 7 for recording information are rotated by a spindle 6 and
Are floating above the surface of the magnetic disk 7 by the slider 5. The slider 5 is connected to the carriage 2 via the head support system 3, and the position of the head 4 with respect to the surface of the magnetic disk 7 is moved by the rotation of the carriage 2 by the VCM 1, thereby recording and reproducing information at a predetermined position. I do. The drive current 15 of the VCM 1 is generated from a position deviation signal (position error signal) 16 obtained from the head 4 by a calculation by a microcomputer (hereinafter, abbreviated as “microcomputer”). That is, the position deviation detector 13,
The head state estimator 12, the seek controller 9, the follow controller 10, the control system switch 8, the initial value setting device 13, and the like are realized by software in a microcomputer.

Hereinafter, the contents of calculation by the microcomputer will be described with reference to FIG.

When the head 4 is moved, the control system switch 8 switches between the seek controller 9 and the follow controller 10 to perform positioning at the target position. This switching is performed by the control system switch 8 detecting that the positional deviation between the current position of the head 4 and the target position has become smaller than a certain value.

In some cases, the seek controller 9, the settling controller (not shown), and the follow controller 10 are switched to perform the positioning at the target position in order to increase the responsiveness of the positioning. In this case, the settling controller determines the parameters with an emphasis on quick response. Since the configuration is the same as that of the follow controller 10, the settling controller can be configured in the same manner as the follow controller 10 described below. When the settling controller is used, the settling controller is changed to the follow controller 10.
Is switched after the head 4 has almost settled to the target position, and the transient response of switching from the settling controller to the follow controller 10 is very small. It may be performed only when switching from the device 9 to the settling controller.

The seek controller 9 is a speed controller for moving the head 4 near a target position. The seek controller 9 adds a head profile 4 to a speed profile 19 indicating the relationship between the head position deviation and the target speed as shown in FIG.
Is controlled so that the speed of the vehicle follows.

The follow controller 10 is a position controller that causes the head 4 to follow a target position, and is configured using a phase lead / lag compensator or a PID controller. In the present embodiment, a first-order phase lead / lag compensator having a small amount of calculation is used. 1
As an example of the next phase advance / delay compensator, the configuration of the compensator shown in Expression (1) is shown in FIG.

C (s) = (αs + ω _C ) / (s + αω _C ) (1)

In the case where digital control (sampling time: Ts) is performed to control the characteristics equivalent to the equation (1), the equation becomes the equation (2) and the configuration is as shown in FIG.

[0018]

(Equation 1)

Similarly, the configuration of the second-order phase lead / lag compensator is shown in FIG.

[0020]

(Equation 2)

Since a residual position deviation cannot be obtained only with the phase lead / lag compensator, it is desirable to connect an integrator.

On the other hand, as a PID controller, a controller based on a speed estimation value, a position estimation value and a position deviation signal by an observer, and feedback from an integrator is considered. The configuration diagram is shown in FIG.

In this figure, the integrator is common to the PID controller and the aforementioned phase lead / lag compensator. Therefore, in the following description, the integrator is omitted for simplification of the expression, and
Consider the PD control shown in FIG. When this diagram is converted into a transfer function expression, it becomes Equation (4).

[0024]

(Equation 3)

This equation is replaced as follows.

[0026]

(Equation 4)

Thus, it is expressed as in equation (6).

C (s) = (b ₂ z ² + b ₁ z + b ₀ ) / (z ² + a ₁ z + a ₀ ) (6)

From this equation, the PD controller and the secondary phase lead / lag compensator are equivalent, and equation (5) is a conversion equation between the PD controller and the secondary phase lead / lag compensator. You can see that it is.

Here, if the gain of the PD controller is selected so that a set of pole zeros of the four pole zeros is canceled by the secondary phase lead / lag compensator, the primary phase lead / lag compensator can be used. Equivalence is clear from equation (3). Therefore, a pole-zero of a frequency having little effect on the first-order phase lead / lag compensator designed to obtain desired characteristics is selected, a second-order phase lead / lag compensator is formed, and the simultaneous equation shown in the equation (5) is obtained. To determine the gain of the PD compensator. In this way, a PD controller equivalent to a first-order phase lead / lag compensator can be created.

Next, as another expression method of the PD controller,
Equation (7) shows the state equation based on the state space model.

[0032]

(Equation 5)

It should be noted that the state quantity is represented by the following equation (8) when represented using the symbols shown in FIG.

[0034]

(Equation 6)

The state equation of the second-order phase lead / lag compensator is given by equation (9).

[0036]

(Equation 7)

Since it is possible to convert each other in the transfer function expression in this equation, it is known that the state equation can be converted as follows by a certain 2-by-3 conversion matrix T.

[0038]

(Equation 8)

From equation (7),

[0040]

(Equation 9)

A transformation matrix T is obtained from Expression (11) from Expressions (9) and (10).

[0042]

(Equation 10)

Accordingly, by performing similarity transformation using the transformation matrix T obtained by the equation (12), the state quantity X of the PID controller is obtained.
[K] is converted into a state quantity W [k] of the second-order phase lead / lag compensator.

W [k] = TX [k] (13)

Further, as a state quantity Q [k] of the first-order phase lead / lag compensator, a state equation is obtained from equation (2) using equation (1).
It is the state quantity at the time of 4).

[0046]

[Equation 11]

Assuming that a transformation matrix for extracting a feedback state quantity that is not canceled by pole / zero is S, Q [k] = SW [k]
= STX [k], the state quantity X [k] becomes Q
Can be converted to [k].

Here, assuming that P _C2 = P _Z2 , S is given by the equation (1)
5).

S = [Kcp _C2 (p _Z1 −p _C1 ) −Kc (p _Z1 −p _C1 )] (15)

As described above, the initial state quantity of the first-order phase lead / lag compensator is directly obtained from the state quantity of the PD controller, that is, the state estimation quantity of the VCM.
An initial value of the state quantity of the compensator suitable for the state quantity of the CM can be obtained, and the settling operation after the seek operation is stabilized.

[0051]

Next, an embodiment of the present invention will be described using numerical examples.

FIG. 1 shows the outline of this embodiment. One or more magnetic disks 7 for recording information are
The spindle 6 rotates. The head 4 is connected to the carriage 2 via a head support system 3, and the position of the head 4 with respect to the disk surface is moved by the rotation of the carriage 2 by the VCM 1.
Recording / reproduction of information is performed at a predetermined position. The drive current 15 of the VCM 1 is generated by a microcomputer based on a position deviation signal detected by the position deviation detector 13 based on a magnetic head signal obtained from the head 4. That is, the position deviation detector 13, the head state estimator 12, the seek controller 9, the follow controller 10, the control system switch 8, the initial value setting device 13, and the like are realized by software in the microcomputer.

Hereinafter, the contents of the calculation by the microcomputer will be described with reference to FIG.

For the movement of the head 4, the controller is switched between the seek controller 9 and the follow controller 10 to perform positioning at the target position. Switching of the controller is performed by the control system switch 8 when the positional deviation between the current position of the head 4 and the target position becomes smaller than a certain value, for example, 1/2 track. The seek controller 9 is a speed controller that moves the head 4 close to a target position. The seek controller 9 controls the head speed to follow a speed profile 19 indicating a relationship between the head deviation and the positional deviation between the current position and the target position of the head 4 as shown in FIG. 3, for example. Actually, the position deviation is obtained from the position deviation signal obtained from the head 4, and the target speed is obtained from a table written in the memory as a relationship between the position deviation and the target speed as shown in FIG. The head speed uses a value estimated by the state estimator 12 based on the actuator model from the drive current of the VCM 1 to the head position / speed.

The follow controller 10 is a position controller that causes the head 4 to follow the target position, and is configured using a phase lead / lag compensator or a PID controller. FIG. 9 shows a frequency characteristic of a first-order phase lead / lag compensator represented by equation (16) as an example of the phase lead / lag compensator.

C (s) = (3s + 3000) / (s + 9000) (16)

If control having the same characteristics is performed by digital control (sampling time: 100 μsec), equation (17) is obtained, and the frequency characteristics are as shown in FIG.

C (s) = 2.079 (z−0.9048) / (z−0.4066) (17)

Since the residual position deviation is not removed only by the phase lead / lag compensator, it is desirable to add an integrator.

In the present invention, as the PID controller, the speed estimation value, the position estimation value, and the position deviation signal by the observer,
Consider a controller based on the feedback of the integrator. The integrator is also necessary for the phase lead / lag compensator as described above.
Since the same quantity may be used as the state quantity, an integrator is omitted in the following description.

The equation is converted to a second-order phase lead / lag compensator by adding a set of poles and zeros. This canceling pole zero depends on the characteristics of the VCM to be controlled, but as an example, 2.5 kHz
Add to

C (s) = 2.079 {(z−0.9048) (z−0.2079)} / {(z−0.4066) (z−0.2079)} = (2.079z ^{2 -2.321z +0.3926) / (z 2 -0.6145z} + 0.08453) ··· (1 8)

From equation (18), the parameters of the PD controller shown in the figure become equivalent when the following values are set.

[0064]

(Equation 12)

Here,

[0066]

(Equation 13)

The transformation matrix T is obtained from the equations (12) and (20).

[0068]

[Equation 14]

Since the transformation matrix S is given by the following equation (21),

S = [0.21689−1.04324] (22)

When switching from the seek controller 9 to the follow controller 10 by converting the estimated value and output value of the observer of the seek controller 9 using the transformation matrix ST in the equation, the initial value of the follow controller 10 can be changed. Is found.

S = [4.94967 × 10 ¹⁴ 1.08080 × 10 ¹¹ 8.2327 0 × 10 ⁶ ] (23)

[0073]

According to the magnetic disk drive of the present invention, when switching from the speed controller to the position controller in the head access operation, the initial state amount of the position controller is obtained from the state amount of the actuator. Variations in the state quantities of the actuator at the time of switching can be corrected by simple calculations.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a magnetic disk drive according to the present invention.

FIG. 2 is a schematic side view showing a head positioning mechanism in the magnetic disk device of FIG.

FIG. 3 is a graph showing an example of a speed profile in the magnetic disk device of FIG. 1;

FIG. 4 is a block diagram showing a continuous-time primary phase lead / lag compensator in the magnetic disk device of FIG. 1;

FIG. 5 is a block diagram of a discrete-time first-order phase lead / lag compensator in the magnetic disk drive of FIG. 1;

FIG. 6 is a block diagram of a discrete-time secondary phase advance / delay compensator in the magnetic disk drive of FIG. 1;

FIG. 7 is a block diagram of a PID compensator in the magnetic disk device of FIG.

FIG. 8 is a block diagram of a PD compensator in the magnetic disk device of FIG. 1;

9A and 9B are graphs showing frequency characteristics of a continuous-time primary phase lead / lag compensator in the magnetic disk device of FIG. 1; FIG. 9A shows frequency-gain characteristics, and FIG. It is a phase characteristic.

10 is a graph showing frequency characteristics of a discrete-time primary phase lead / lag compensator in the magnetic disk device of FIG. 1; FIG. 10 (A) shows frequency-gain characteristics; FIG.
(B) is a frequency-phase characteristic.

[Explanation of symbols]

 Reference Signs List 1 VCM (actuator) 2 Carriage 3 Head support system 4 Head (magnetic head) 5 Slider 6 Spindle 7 Magnetic disk 8 Control system switch 9 Seek controller (Speed controller) 10 Follow controller (Position controller) 11 Initial value Setter 12 Head state estimator 13 Position deviation detector 14 Control system switching signal 15 Drive current 16 Position deviation signal 17 Magnetic head signal 18 Estimated state quantity 19 Speed profile

Claims (3)

[Claims] 1. A head positioning mechanism having an actuator for moving a head for recording and reproducing data on and from a magnetic disk, and a position for detecting a positional deviation between a current position and a target position of the head via the head. A deviation detector; a head state estimator for estimating a head estimation speed and a head estimation position based on the position deviation; and a drive signal output to the actuator based on the position deviation to move the head close to a target position. A speed controller for causing the head to follow a target position by outputting a drive signal to the actuator based on the position deviation, and a position controller comprising a first-order phase lead / lag compensator; A control system switch for switching and operating the speed controller and the position controller based on the speed; When switching from the degree controller to the position controller,
A magnetic disk drive comprising: an initial value setting unit that sets a state quantity of the position controller based on the drive signal, the position deviation, the estimated head speed, and the estimated head position. 2. The method according to claim 1, wherein the actuator is a voice coil.
2. The magnetic disk drive according to claim 1, wherein the magnetic disk drive is a motor. 3. An integrator is provided in said position controller,
The magnetic disk drive according to claim 1.