JPH09219074A - Magnetic disk device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To shorten a seek time by utilizing the seeking ability of a device as much as possible while allowing that the seed reducing current to a voice coil motor is saturated. SOLUTION: This device is constituted so that the changeover from an acceleration mode to a speed reduction mode is performed at the time when a remaining distance reaches a certain value Xs (m) in which the remaining distance X (m) is preliminarily made to be a fixed equation or a table with respect to a total moving distance Xa (m) in an each seek control. The acceleration command value with respect to the voice coil motor is controlled by issuing an acceleration command value larger than the acceleration capable of being generated at its maximum in an acceleration mode and by issuing the speed reduction acceleration command value I (m/s<2> ) determined an operation being I (m/s<2> )=C×V<2> /X (C = a fixed coefficient 0.5-1.0) by using the remaining distance X (m) to a target track and an actual velocity V (m/s) to be obtained every sampling cycle in a speed reduction mode.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to positioning control of a magnetic disk device.

[0002]

2. Description of the Related Art A conventional control method relating to actuator movement (hereinafter referred to as seek) control of a magnetic disk device is as follows.
The target speed Vguide (m / s) corresponding to the remaining distance X (m) to the target track is calculated, and the deviation Verr (= Vguid) from the actual moving speed V (m / s) of the actuator is calculated.
A method of calculating e-V) and performing feedback control so that Verr = 0 is widely used. That is, this is a method of controlling the movement to the target track by setting one target trajectory on the phase plane (X-V plane) and following the target trajectory without deviation.

As a prerequisite for realizing such seek control, it is necessary to set a target velocity trajectory that can be followed under various conditions such as power supply voltage fluctuations. In other words, in the deceleration control region where the deviation to the target trajectory is hardly allowed, the target trajectory that can be followed in the range where the voice coil motor current is not saturated must be set, and the true seek capability of the device under normal conditions is set. Is a control method that reduces the.

As a method for compensating for such a drawback, Japanese Patent Application Laid-Open No. 5-89613 has filed a patent that control can be performed even when the deceleration current is saturated by setting a timing at which the acceleration capability of acceleration is switched to deceleration. However, it is considered to be an extension of the conventional technique in the concept of finally following one target trajectory.

[0005]

The performance required for seek control as a magnetic disk device is to complete the movement of a predetermined distance within a predetermined time. There are innumerable orbits on the phase plane that satisfy this requirement under normal equipment use environment, but seek control by the conventional technology is the most seek time that satisfies the minimum seek time specification from the infinite number of orbits. We have only selected one trajectory that requires

An object of the present invention is to reduce seek time by providing a seek control method which allows saturation of deceleration current and utilizes the seek capability of the apparatus to the limit.

[0007]

Needless to say, the shortest seek control in a magnetic disk device is seek control by saturation acceleration and saturation deceleration, but only during seek operation.
It is indispensable to accurately grasp the switching timing Ts from acceleration to deceleration, which comes only once, and it is impossible to realize it.

However, proposing a control method that approaches the saturation acceleration / deceleration control as much as possible is a means for solving this problem, and the items to be clarified for that purpose are:
There are two points below.

Setting of conditions for ending the saturated acceleration mode.

Generation of a command value on the assumption of current saturation to the voice coil motor in the deceleration mode.

With regard to (1), the acceleration is terminated when the remaining distance X (m) reaches a certain value Xs (m) which is previously formulated or tabulated with respect to the total movement distance Xa (m) in each seek control. As a condition.

With respect to the above, the concept of setting one target trajectory on the phase plane is abandoned, and the remaining distance X (m) to the target track and the actual speed V obtained at each sampling cycle of the DSP for performing seek control. The acceleration command value I (m / s ² ) from (m / s) to the voice coil motor is displayed on the DSP.
It can be realized by directly determining by the calculation of.

By implementing the present invention, it is possible to perform control that allows the deceleration current flowing through the voice coil motor to be saturated in the deceleration region during seek control, and as a result, the seek time can be shortened. .

[0014]

1 is a block diagram showing the configuration of the entire magnetic disk device including a seek controller according to the present invention. The positioning control method of the magnetic disk device is roughly classified into a servo surface servo method having a dedicated servo disk for positioning and a data surface servo method having periodically servo information in the data disk. However, the latter data surface servo method will be described as an example.

The entire apparatus includes a read / write control section for writing data on the disk surface and reading data written on the disk surface, and a positioning apparatus for moving the head to an arbitrary track in the radial direction for positioning. It can be roughly divided into control units.

The positioning control section includes a position detection circuit for extracting servo information previously written on the disk from the signal read from the head, and an A / D for discretizing the position signal generated by the circuit. A conversion unit, a controller unit by a DSP for calculating and deriving an operation amount to be applied to the voice coil motor based on the discretized position signal, a D / A conversion unit for converting the calculation result into a continuous current amount, and A voice coil motor unit that generates a force proportional to the generated electric current and a carriage that displaces the head in the radial direction by the generated force form a loop feedback loop.

The positioning control unit is a controller D
It can be classified into a seek control unit and a following control unit inside the SP, and the seek control unit according to the present invention calculates a voice coil motor current command value in the acceleration mode and a mode determination unit that determines the acceleration mode or deceleration mode during seek. And a deceleration command value calculation unit that calculates a voice coil motor current command value in the deceleration mode.

An example of seek control according to the present invention will be described below.

In the acceleration mode, as described above, the remaining distance X (m) is set to a certain value Xs (m) which is previously formulated or tabulated with respect to the total movement distance Xa (m) in each seek control. The condition is to end acceleration when it reaches. A certain value Xs (m) is Xa when the counter electromotive voltage generated in the voice coil motor in the seek operation is added.
It can be qualitatively and easily imagined that the value will be smaller than / 2, and quantitatively it should be calculated in advance by simulation or the like. In the acceleration control region, an acceleration command value larger than the maximum acceleration that can be generated for the voice coil motor is issued as a current command value.

Prior to description of the deceleration mode, deceleration control during seek according to the conventional technique will be described. FIG. 6 is a phase plan view showing the relationship between the moving position (distance to the target track) of the head and the moving speed when the deceleration control according to the conventional technique is performed. In the conventional seek control, one target trajectory is set in advance on the phase plane, and the starting point A
Feedback control is performed to move from (Xa, Va) to the end point B (Xb, Vb) along the set trajectory V = f (X).

Specifically, as shown in the block diagram of FIG. 7, the seek control during deceleration is performed by a block for generating a reference speed command corresponding to the remaining distance to the target and the actual moving speed of the head. Is comprised of a speed detection section for detecting the speed deviation, an error amplifier section for amplifying the speed deviation, and a feedforward section for previously compensating for the calculated deceleration current necessary for deceleration to minimize the speed deviation. Due to the loop feedback loop in FIG. 7, the trajectory actually drawn by the head substantially matches the target trajectory.

On the other hand, the time T (A → B) required to move between two points on the phase plane is dX using the trajectory f (X).
From / dT = f (X),

[0023]

[Equation 1]

The farther the trajectory f (X) draws from the X axis, the shorter the time T required to move between the two points.

However, the target velocity trajectory f0 (X) set in the prior art is set on the premise that the carriage on which the actual head is mounted can follow under various environmental conditions such as power supply voltage fluctuations. Therefore, the speed is controlled at a speed with a margin larger than the marginal speed such as the limit trajectory shown in the figure. It should be possible to draw a trajectory of the region that has been traveled with a shorter travel time.

The seek deceleration control method according to the present invention will be described below.

The performance required for the deceleration control is the moving speed V = 0 when the target track (X = 0) is reached, and in addition, from the viewpoint of continuous control, it is desirable that the performance is when the target track (X = 0) is reached. This means that deceleration (= acceleration current) = 0. Therefore, here, the deceleration control start time A
From (Xa, Va), let us consider a control method in which any of position, velocity, and acceleration continuously converges to zero.

In the first control method according to the present invention, three of position, velocity and acceleration are indexed (EXP) in the time domain.
A method of attenuating according to a function is used. Initial conditions (Xa,
Using Va), the three relationships are: X (t) = EXP (−k × t) × Xa V (t) = − k × EXP (−k × t) × Xa I (t) = k ² × EXP (−k × t) × Xa ... (2) In order to realize the first method, the formula (2) is integrated using the position / velocity (X, V) obtained at each sampling timing discretized in the deceleration control, and the deceleration I = V ² / X ... It is sufficient to issue the amount determined by the equation (3) as the voice coil motor current command value. As shown in FIG. 2, the control realized by this method is a trajectory fn (X) that linearly connects the state (Xn, Vn) at the sampling time Tn on the phase plane and the origin (0, 0).

However, the point that this first control is decisively different from the conventional control is that, with respect to the trajectory fn (X) set at the time Tn, the next sampling time Tn + due to the deceleration current saturation.
Even if it is impossible to follow the trajectory fn (X) in 1, the new trajectory fn + 1 is set and the control is continued. Finally, the transition from the time Tm when the intersection of the tangent of the trajectory drawn on the phase plane and the Y axis to the negative region from the origin to the negative control shifts to unsaturated control, and the trajectory fm
(X) linearly converges on the origin. That is, the trajectory fm
Up to (X), the deceleration current is in the saturated region, and thereafter, the deceleration current is in the non-saturated region.

The necessary condition for realizing this control is that the intersection point between the trajectory drawn on the phase plane and the X axis in the deceleration current saturation state is in the positive region, that is, speed = 0 is guaranteed when the target track is reached. There is a deceleration margin to
This depends only on the timing of switching from acceleration to deceleration. FIG. 3 shows the simulation result of seek control by this control method. According to FIG. 3, it is understood that the deceleration current is attenuated according to the EXP function and finally becomes 0 after shifting to the unsaturated state in the deceleration control.

As the first control method, the convergence according to the EXP function has been described as an example. In principle, however, the movement along the EXP function requires ∞ time to reach the target track (Fig. (Infinite time is required for the current at 3 to finally converge to 0). As a means for avoiding this problem, there is a practical technique such as slightly offsetting the final target position / velocity (X, V) in seek control from the origin, but here another control method will be considered.

As a second control method, let us consider a method in which the deceleration current is proportionally attenuated according to a linear function in the time domain and three of position, velocity and acceleration converge to 0 when the target track is reached. At time t = 0 when the target track is reached, the relationship between position X, velocity V, and acceleration I is as follows: I (t) = k × t V (t) = (1/2) × k × t ² X (t) = (1/6) × k × t ³ Equation (4) is obtained. In order to realize this second control method, the formula (4) is integrated using the position / velocity (X, V) obtained at each sampling timing discretized in the deceleration control, and the deceleration acceleration is obtained. I = (2/3) × V ² / X ... It is only necessary to issue the amount determined by the equation (5) as the voice coil motor current command value. FIG. 4 shows the simulation result of seek control by this control method. According to FIG. 4, the deceleration current is 1 after shifting to the unsaturated state in the deceleration control.
It can be seen that deceleration follows the next function.

Further, as a third control method, the deceleration current is a constant value, and when the target track is reached, two of the position and speed are zero.
Consider how to converge to. Assuming that time t = 0 when the target track is reached, the relationship between position, velocity and acceleration is as follows: I (t) = k V (t) = k × t X (t) = (1/2) × k × t ² ... Equation (6) is obtained. In order to realize the third control method, the formula (6) is integrated by using the position / velocity (X, V) obtained at each sampling timing discretized in the deceleration control to calculate the deceleration acceleration. I = (1/2) × V ² / X ... It is sufficient to issue the amount determined by the equation (7) as the voice coil motor current command value. FIG. 5 shows a simulation result of seek control by this control method. According to FIG. 5, it can be seen that the deceleration current in deceleration control has a constant value.

As described above, three control methods which allow the saturation of the voice coil motor current in the deceleration control have been considered. The expressions (3), (5) and (7) are all obtained at the sampling timing.・ Deceleration acceleration I = C × V ² / X (C = constant coefficient) using speed (X, V) ... The amount determined by equation (8) may be issued as a voice coil motor current command value. I came to the conclusion. Where constant coefficient C
Is a typical function form of the deceleration current at a value of C = 1 or (2/3) or (1/2), but is continuous in the range of C = 0.5 to 1.0. You can easily imagine having. Further, the determination of the voice coil motor current command value by this control method is very simple and can be easily realized as compared with the control system which is constituted by the generation of the target velocity trajectory according to the prior art.

[0035]

According to the present invention, the seek time can be shortened.

Further, according to the present invention, the memory of the control microprogram and the number of execution steps can be greatly reduced, and the load on the microprocessor can be reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an entire magnetic disk device according to the present invention.

FIG. 2 is a phase plan view when deceleration control is performed according to the present invention.

FIG. 3 shows a simulation result of seek control according to the present invention.

FIG. 4 shows a simulation result of seek control according to the present invention.

FIG. 5 shows simulation results of seek control according to the present invention.

FIG. 6 is a phase plan view when performing deceleration control according to a conventional technique.

FIG. 7 shows a block diagram of a configuration of a deceleration control system according to a conventional technique.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Takashi Yamaguchi 2880 Kozu, Odawara City, Kanagawa Pref., Storage Systems Division, Hitachi, Ltd.

Claims (6)

[Claims] 1. To position an actuator having a magnetic head on an arbitrary target track, a current is supplied to a voice coil motor to generate an acceleration proportional to the current, and then a reverse current is supplied to decelerate the actuator. And a remaining distance X to a target track obtained at each sampling timing in a deceleration control region during the movement control.
(M) and the detected value of the moving speed V (m / s) of the actuator, the deceleration acceleration command value I for the voice coil motor
(M / s ² ) is determined for each timing, and a current is supplied to the voice coil motor based on the deceleration acceleration command value I, and the deceleration acceleration command value I exceeds the saturation value of the voice coil motor current. A magnetic disk drive characterized by acceptable control. 2. The deceleration acceleration command value I according to claim 1, wherein I (m / s ² ) = C ×, where X is a remaining distance to a target track, V is a moving speed of an actuator, and C is a constant coefficient. A magnetic disk device characterized by being defined by V ² / X. 3. The deceleration acceleration command value I according to claim 1, wherein when the deceleration acceleration command value I exceeds a saturation value of the voice coil motor current, the voice coil motor is controlled by the saturation value. Is smaller than the saturation value of the voice coil motor current,
A magnetic disk device, wherein the voice coil motor is controlled based on the deceleration acceleration command value I. 4. In order to position an actuator equipped with a magnetic head on an arbitrary target track, a current is supplied to a voice coil motor to generate an acceleration proportional to the current, and then a reverse current is supplied to decelerate the actuator. A magnetic disk device for performing movement control by means of an acceleration control mode and a deceleration control mode, the acceleration being greater than the maximum acceleration that can be generated for the voice coil motor in the acceleration control region. The command value is output as the current command value, and the remaining distance X to the target track in the deceleration control area
(M), the moving speed V (m / s), and the constant coefficient C, I
A magnetic disk device characterized by outputting as a deceleration acceleration command value for a voice coil motor based on an operation defined by (m / s ² ) = C × V ² / X. 5. The control according to claim 4, wherein the transition from the acceleration control mode to the deceleration control mode is performed when the remaining distance to the target track becomes smaller than half the total travel distance. A magnetic disk device characterized by the above. 6. The magnetic device according to claim 2, 3, 4 or 5, wherein the constant coefficient C in the deceleration control region is set to a value of 0.5 or more and 1.0 or less. Disk device.