JPH11120720A - Method and device for positioning head of magnetic disk device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To correct the waviness of a head position information signal so as to improve adjacent track pitch accuracy by making and recording the correction data of the rotational synchronization component of a head position information signal in a servo information area, reading the correction data at the time of head positioning to correct a real head position difference signal and performing head positioning by using the position difference signal. SOLUTION: In a run-out processing part 2 included in a positioning control unit 1, a waviness component synchronized with disk rotation is calculated for each sector by an arithmetic operation using a Fourier transformation expression. The run-out processing part 2 measures a real head position difference amount by one rotation with an index on a disk as a reference, calculates each correction coefficient and stores in a memory provided therein. After the measuring, a correction value is added to a real head position difference signal U1 used for normal head positioning by an adding/subtracting part 4 to generate a head position difference signal U2, and then normal head positioning is performed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a head positioning control method and apparatus for a magnetic disk drive constituted by a closed loop head positioning control system.

[0002]

2. Description of the Related Art In a magnetic disk drive of closed loop control, a head position information signal for positioning a head at a desired track is, as shown in JP-A-58-222468, a burst of an even track and an odd track. It is generally known to be generated from the amplitude difference of

The recording pattern of the head position information signal is called a servo pattern and is recorded by a servo track writer (STW) before product shipment, and is not rewritten after recording.

The head position information signal is information serving as a reference for positioning the head on a desired track. It is very important that the head position information signal is accurately recorded in a high track density (high TPI) magnetic disk device. Is important.

In order to increase the track density of the magnetic disk drive in recent years, it is necessary to improve the accuracy of the adjacent track pitch, that is, the accuracy of the head position information signal.

Conventional techniques for improving the accuracy of a head position information signal are disclosed in JP-A-1-279474, "Head positioning method", JP-A-4-324173, "Magnetic disk device positioning control method", 7-24
No. 9276, "Off-track correction circuit of magnetic disk device", and JP-A-6-176514, "Disk device".

[0007]

In a magnetic disk drive, a head position information signal and a head position information signal are discretely provided between a plurality of divided data areas in order to improve a data format efficiency while improving a high precision head positioning technique. The current mainstream is a sector servo system in which a so-called servo information signal is arranged. However, since the number of servo sectors is directly related to the data format efficiency, it is necessary to reduce the number of servo sectors, that is, to lower the sample rate.

As the sample rate decreases, it becomes difficult to increase the control band for positioning the magnetic head.
The compression ratio characteristic of the control system in the low frequency region will be deteriorated.

Further, the head position information signal of the magnetic disk drive of the sector servo system is sufficiently flat due to radial vibrations due to eccentricity of the center axis of the disk and fluctuations due to the location of the reproduced signal. Not necessarily.

As a result, in the conventional magnetic disk drive,
Since the head also vibrates in response to the undulation of the head position information signal, track pitch fluctuation, so-called run-out occurs. Among these run-out components, the primary run-out component (fundamental wave component) particularly at the rotation frequency of the disk is large, and this primary component is usually in a low frequency region around 90 Hz. For this reason, as the sample rate decreases,
If the compression ratio characteristic of the control system in the low frequency region is deteriorated, it is difficult to secure the positioning accuracy of the head (head following characteristic) due to the influence of the runout.

As described above, when the head follow-up characteristic is deteriorated due to the effect of the run-out when the compression ratio characteristic is reduced, the S / N of the data reproduction signal is reduced, and a high positioning accuracy cannot be obtained in realizing a high TPI.

Therefore, in the magnetic disk drive, it is very important to reduce the runout generated in the low frequency region in synchronization with the rotation of the disk, in order to satisfy the required device specifications. .

None of the above publications disclose reduction of runout components.

Accordingly, an object of the present invention is to provide a head positioning control method and apparatus which solves the above problems.

[0015]

According to the present invention, there is provided a magnetic disk drive head positioning control method and apparatus in which a disk vibrates in a radial direction due to eccentricity of a disk center axis or the like.
The undulation of the head position information signal resulting from a phenomenon called runout generated thereby is corrected, and the accuracy of the adjacent track pitch is improved.

For this purpose, first, a rotation synchronization error correction value which is correction data of a rotation synchronization component of a head position information signal is created, and this correction data is recorded in a servo information area of a disk.

At the time of actual head positioning, a rotation synchronization error correction value recorded in the servo information area is read and added to an actual position error signal to correct a position error due to undulation of the disk. The head is positioned to a desired track position using the position error signal thus obtained.

In creating the rotation synchronization error correction value,
Since only the primary run-out component is extracted from the position error signal using the Fourier series relational expression and the correction value is calculated, the primary run-out component can be reliably reduced.

Further, according to the positioning control method of the present invention, the result of measuring the head position error amount over one round is stored in the memory and added to the actual head position error signal after the end of the measurement in the rotation cycle. This is regarded as a head position error signal and positioning is performed.

At this time, the measurement result is added as a feedforward component, and the head performs a following operation determined by the compression ratio characteristic with respect to the position error signal of the original control system.

If the original control system can follow the target value given by the feed forward with a residual position error amount of about 1/10, it is compressed by about 1/10 from the initial position error amount. looks like. When this is measured again and added, another 1/10, that is, the first 1/1
This means that the data has been compressed to about 00. This is the principle of the run-out correction in the present invention, and is the premise of the correction data to be written.

[0022]

FIG. 1 is a block diagram showing a head positioning control device according to the present invention in a magnetic disk drive using a sector servo system.

This positioning control device includes a magnetic head 10
The positioning control unit 1 supplies a VCM drive current to a voice coil motor (VCM) 8 that drives an actuator mechanism 11 that performs positioning.

The positioning control unit 1 has a microcomputer 12, which is shown in a functional block diagram in the figure. The microcomputer 12 includes a runout processing unit 2, addition and subtraction units 3 and 4, and a compensation unit 5. As shown in FIG. 1, the positioning control unit 1 includes an analog-to-digital (A / D) converter 9 that receives the output S1 of the magnetic head 10 as an input, and a digital analog (D) converter that receives the control amount output from the compensating unit 5 as an input. / A) A converter 6 and an amplifier 7 are provided.

Through the A / D converter 9 which performs analog / digital conversion, a discrete head position information signal S2 unique to the sector servo system is obtained.
Based on the head position information signal S2, a rotation synchronization error correction value in each servo sector is calculated and output to the host interface circuit 13.

The upper interface circuit 13 records the rotation synchronization error correction value at a predetermined location provided in the sector servo area of the magnetic disk as shown in FIGS. FIG. 2 shows a sector servo area 21 and a data area 22 of the magnetic disk 20, and FIG. 3 is an enlarged view showing a part of the sector servo area and the data area.

The compensator 5 determines an appropriate control amount required for driving the voice coil motor VCM 8 through the digital / analog converter (D / A) 6 based on the position error signal U 2 to which the read rotation synchronization error correction value is added. The amplifier 7 outputs the VCM drive current value S3 for moving the head 10 to a desired track position to the VCM 8.

FIG. 3 shows an example of a pattern in which correction data is recorded in a servo information area of a magnetic disk. FIG.
In the figure, SM is a servo sector synchronization area, G is a gray code area, and BP is a burst signal area. The approximate position of a track is detected, and high-resolution head position information is detected in the burst signal area of the BP. In the present embodiment, a correction data storage area is provided as RC after this area,
The correction value (correction data) for each sector is stored.

This correction data is obtained by a method described below, and then coded to perform gray code detection.
It is recorded so that it can be detected by the / D converter 9. Although not shown in FIG. 1, recording is performed by ordinary writing means.

Next, the operation will be described. First, a method of creating correction data and recording it on a disc will be described.

The run-out processing unit 2 in the positioning control unit 1 in FIG. 1 calculates a swell component (run-out component) for each sector in synchronization with the rotation of the disk by an operation using a Fourier transform equation. A method of extracting a swell component synchronized with the rotation cycle of the disk, particularly a primary run-out will be described below.

The first-order component and the DC component of the runout are obtained using a general Fourier series. In general, it has been proved that the coefficient value of the Fourier series expansion is an approximation that minimizes the mean square of the error. I have.

Signals can be classified into periodic signals and non-periodic signals from the viewpoint of periodicity. A periodic signal repeats the same waveform at a certain period. If the period of the signal X (t) is P,

[0034]

X (t) = X (t + np) (1) Therefore, to describe a periodic signal, the signal value between any time t ₀ and one cycle

[0035]

It is sufficient to provide X (t), t ₀ ≦ t <t ₀ + P (2), and the periodic signal can be expanded into a Fourier series. Here, the runout is also a periodic signal having a certain period P except for the asynchronous component. If the primary component of the runout is represented by x (t), x (t) is expressed by the sum of a constant term and a sine wave signal having a period P. It can be approximated as

[0036]

X (t) ≒ a ₀ + a ₁ cos ωt + b ₁ sin ωt (3) where each coefficient is as follows.

[0037]

(Equation 4)

Here, ω is the frequency of the primary run-out component, that is, the fundamental rotation frequency, and ω = 2
π / P. N is the number of servo sectors per one round of the data track, and ωtk is 2π in one rotation based on the index.

The values of sin and cos are stored in a table, and these values are counted from the index to identify the number, and the table is searched.

Furthermore, the approximation accuracy can be improved by including the second, third,..., Nth harmonic components. In theory, the signal x (t) should be completely matched with n → ∞. In this case, it is expressed by the following equation.

[0041]

(Equation 5)

Next, a procedure for creating a correction value in the run-out processing section 2 will be described with reference to the flowchart of FIG.

[Procedure 1] The run-out processing unit 2 measures the actual head position error amount for one rotation based on the index on the disk (step S1), and formulas (3) to (6).
Are used to calculate the respective correction coefficients a ₀ , a ₁ , and b ₁ and store them in the memory in the run-out processing unit 2 (step S2).

[Procedure 2] Immediately after the completion of the measurement, an actual head position error signal U1 used for normal head positioning is used.
(Obtained by adding the head position information signal S2 to the target value input r (t) by the adder / subtractor 3)
Then, the head position error signal U2 is generated by adding the correction value x (t) obtained by the equation (3) (step S3), and normal head positioning control is performed (step S4).

[Procedure 3] In consideration of the convergence of the transient response due to the addition of the correction value, the process waits for 1/3 rotation (step S5). However, the waiting time is set arbitrarily.

[Procedure 4] The second measurement is started (step S6). However, in this case, unlike the first measurement,
Since the waiting time is taken into consideration, the measurement start position is not always the index, so it is necessary to pay attention to table drawing.

[Procedure 5] After the second measurement, each correction coefficient stored in the memory is updated and stored again (step S7).

[Procedure 6] Similarly, the third measurement and updating of each correction coefficient are performed (step S8).

[Procedure 7] Thereafter, if the residual position error is equal to or less than an arbitrary set value (step S9), the measurement is terminated (step S10). If not, the same measurement and update are repeated (step S11).

An example of a calculation algorithm will be described below.

[0051] <1 time for the measurement> a ₀ = ₀ each coefficient initial _{value: a 1 = b 1 = a} 3 = b 3 = 0 (8) Rcyl_smpl (t k) = x (t k): in t _k off-track amount (remaining seek length) sect_nm: N = 60 represents what number of sectors from the index: total number of servo sectors per revolution data track _{a 1 = a 1 + Rcyl_smpl (} t k) · cos (2π · sect_nm / N (9) b ₁ = b ₁ + Rcyl_smpl (t _k ) · sin (2π · sect_nm / N) (10) These relations are calculated for one cycle based on the index.
That is, the process is repeated N times. After repeating for one cycle, the values of Expressions (4) to (6) are obtained. Here, it should be noted that the obtained a ₁ is rewritten as a ₁ data.

Correction value initial value: a ₁ data = b ₁ data = a ₃ data = b ₃ data a = 0 (11) a ₁ data = a ₁ data + (2a ₁ / N) (12) b ₁ data = b ₁ data + (2b ₁ / N) (13) Each coefficient is obtained by the above series of calculations.

<First Correction> The first correction is performed using the relationship of equation (3). A correction value x (t) is added to the actual position error signal, and control is performed by regarding this as a position error signal. The following relational expression is used.

Pa (t _k ): actual remaining seek length Pa (t _k ) = Pa (t _k ) + a ₁ data · cos (2π · sect_nm / N) + b ₁ data · sin (2π · sect_nm / N) (14) <Regarding the Second and Subsequent Measurements and Corrections> The same procedure is repeated. However, it should be noted that the new coefficient value obtained in the second measurement is updated after being added to the coefficient value obtained in the first measurement.

After the run-out processing section 2 calculates the correction coefficient as described above, the run-out processing section 2 outputs the correction coefficient to the upper interface circuit 3, and the upper interface circuit 3 converts the series of data shown in FIG. Is written to a predetermined position in the sector servo information area 21 shown in FIG.

During the actual head positioning operation, the correction data is read simultaneously with the head position information in the sector servo area, and a correction value x (t) is obtained and added to the actual head position error signal U1, thereby obtaining the head position error signal U2.
Is generated, and the head is positioned to a desired track by the operation of the compensating unit 5.

[0057]

The first effect is that it is possible to correct the undulation of the head position information signal caused by the runout caused by the eccentricity of the center axis of the disk of the magnetic disk device, and to improve the accuracy of the adjacent track pitch.

The reason is that the correction data is coded and recorded for each sector in the servo information area required for head positioning, so that it is detected by the same detection circuit as the circuit for detecting the gray code. This is because an appropriate head position error signal corresponding to the undulation of the disk can be generated instantaneously.

A second advantage is that no special hardware is required. Therefore, the present invention can be applied to a small magnetic disk drive.

The reason for this is that a series of calculations and operations relating to the creation of the important correction data and the actual correction in the present invention are all processed inside the F / W (firmware). Further, since a series of processing is performed inside the firmware, efficient processing can be performed without lowering the data format efficiency.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a head positioning control device of the present invention.

FIG. 2 is a diagram showing a sector servo area and a data area of a magnetic disk.

FIG. 3 is a diagram showing an example of a pattern in which correction data is recorded in a sector servo information area of a magnetic disk according to the present invention.

FIG. 4 is a flowchart illustrating a procedure for creating a correction coefficient in a run-out processing unit.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Positioning control part 2 Runout processing part 3, 4 Addition / subtraction part 5 Compensation part 6 D / A converter 7 Amplifier 8 Voice coil motor 9 A / D converter 10 Magnetic head 11 Actuator mechanism 12 Microcomputer 13 Upper interface circuit 20 Magnetic disk 21 Sector servo area 22 Data area

Claims (6)

[Claims] 1. A sector servo type magnetic disk drive comprising a closed loop head positioning control system.
What is claimed is: 1. A head positioning control method for reducing a runout component of a head position information signal arranged in a servo sector on a disk, which is caused by undulation of the disk, comprising: generating correction data of a rotation synchronization component of the head position information signal; Is recorded in a servo information area on a disk, and at the time of actual head positioning, the correction data is read to correct an actual head position error signal, and the head is positioned using the corrected position error signal. Positioning control method for a magnetic disk drive. 2. A case of the reduced as the run-out component of the first-order component, the correction data, the run-out primary component x (t), the Fourier series _{x (t) ≒ a 0 +} a 1 cosωt + b 1 sinωt ( However ω head positioning control method according to claim 1, characterized in that the value of the coefficient a _0, a _1, b ₁ in the case of approximating the frequency) of the run-out first order component. 3. The actual head position error amount is measured for one rotation of the magnetic head, the coefficient is obtained from the measured position error amount and stored. The component x (t) is added to obtain a head position error signal, the position of the magnetic head is controlled based on the position error signal, and the measurement is repeated until the residual position error becomes equal to or less than a predetermined value to update the coefficient. 3. The method according to claim 2, wherein
The head positioning control method according to the above. 4. A sector servo type magnetic disk drive comprising a closed loop head positioning control system, means for creating correction data for a rotational synchronization component of a head position information signal, and storing the correction data in a servo information area on the disk. Means for recording, and at the time of actual head positioning, means for reading out the correction data to correct the actual head position error signal, and for positioning the head using the corrected position error signal. A head positioning control device characterized in that a run-out component of a placed head position information signal caused by undulation of a disk is reduced. 5. A case of the reduced as the run-out component of the first-order component, the correction data, the run-out primary component x (t), the Fourier series _{x (t) ≒ a 0 +} a 1 cosωt + b 1 sinωt ( 5. The head positioning control device according to claim 4, wherein ω is a _value of coefficients a ₀ , a ₁ , and b ₁ when approximated by a runout (first-order component frequency). 6. An actual head position error amount is measured for one rotation of the magnetic head, the coefficient is obtained from the measured position error amount and stored, and the run-out primary degree relating the actual head position error amount to the coefficient is calculated. The component x (t) is added to obtain a head position error signal, the positioning of the magnetic head is controlled based on the position error signal, and the above error is repeated until the residual position error becomes equal to or less than a predetermined value to update the coefficient. 6. The method according to claim 5, wherein
A head positioning control device as described in the above.