JPH04258808A - Magnetic disk device - Google Patents

Abstract

PURPOSE:To attain an accurate positioning by providing a common synchronizing bit to each track and the positioning bits having magnetizing inversion at the different positions in the peripheral direction for each track, and making the positions of positioning bits corresponding to each track to be respectively different for different cells. CONSTITUTION:A positioning pattern 204 is divided into plural groups 205 and each group is constituted of two cells 206. One synchronizing bit 221 and the positioning bits A222, B223 are provided in each cell. Centers of recording widths on each bit are deviated by 1/2 track portion against the center of track. The magnetizing inversion exists inevitably in the synchronizing bit 221 and also exists in either positioning bit A222 or B223. Then, the arrangement of magnetizing inversion in the positioning bits is repeated for every two tracks, and the front and rear positions of bits A222, B223 in regard to the peripheral direction of magnetizing inversion become alternate for every other cell. Thus, an influence by external disturbance is reduced and the accuracy is improved.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for recording patterns for positioning actuators in magnetic disk drives.

0002

[Background Art] Magnetic disk drives, which are the main external storage devices for computers, have become larger in capacity and faster, and the control of actuators that control the movement of magnetic heads has also become more sophisticated. A more compact and highly accurate pattern is also required.

In a position servo system that causes a magnetic head of a magnetic disk device to follow a target track, a position error signal indicating the position of the magnetic head is created using a positioning pattern written on a magnetic disk. A conventional example of a positioning pattern is the Dibit pattern described in U.S. Pat. No. 3,534,344.
ern), the tribit pattern (Tribit pattern) described in U.S. Pat.
Pattern), as well as U.S. Patents 4,238,809 and 46
The modified dibit pattern shown in No. 31606 and the burst format, which is the simplest structure, were described in US Pat. No. 3,185,972 and US Pat. Further, the above-mentioned positioning pattern has been used in both a surface servo method in which one surface of the magnetic disk is used as a dedicated servo surface and a sector servo method in which a part of the sector of each data surface is used as a servo section.

The minimum length of a positioning pattern unit (hereinafter referred to as a "cell") necessary to create a position error signal will be explained. FIG. 8 is a detailed diagram showing a cell with a conventional dibit pattern, FIG. 9 is a detailed diagram showing a cell with a conventional tribit pattern, and FIG. 10 is a detailed diagram showing a cell with a conventional modified dibit pattern. Figure 8, Figure 9, Figure 10
The positioning pattern shown in is a magnetic disk (
This is a partial development of a magnetization pattern formed concentrically on a surface (hereinafter referred to simply as a "disk" for simplicity), and for convenience, the fan angle is ignored and the radial direction is drawn by parallel vertical lines. Therefore, the disk rotation direction is arrow 10.
5, and 106 in the figure indicates the position of the magnetic head (more precisely, the magnetic head gap position, but to avoid any misunderstanding, it will be simply referred to as the "head" below), and arrow 107 indicates the disk radial direction. move along. Head 106 is positioned at the centerline 102, 103, 104 of each track. Further, the solid line in the figure indicates the magnetization reversal position.

[0005] Here, the minimum positioning pattern units, ie, cells, necessary for creating a position error signal are shown by broken lines. Further, assuming that the interval between one magnetization reversal is 1 bit, the cell length of the dibit pattern cell 108 in FIG. 8 is 4 bits. The cell length of the tri-bit pattern cell 109 in FIG. 9 is 3 bits. The cell length of the modified dibit pattern cell 110 in FIG. 10 is 6 bits. In the modified dibit pattern, two additional bits are often added to the synchronization bits 111 and 112, and in this case, the number of synchronization bits is 8 bits.

[0006]

Problems to be Solved by the Invention The long cell length of the positioning pattern has had an adverse effect on magnetic disk drives from the following points. In the case of the surface servo method, the number of cells that can be placed in one track decreases, so the number of samplings used to create a position error signal decreases, and positioning accuracy deteriorates. On the other hand, in the case of the sector servo system, this effect is even more remarkable, and since the positioning pattern can occupy only a few percent of one track, the number of cells in the servo section in one sector is from several cells to at most 10 cells or less. Therefore, in dibits and modified dibits with long cell lengths, the number of samplings is small, the accuracy and reliability of the position error signal is low, and the following accuracy during recording and reproduction deteriorates, which hinders the increase in the capacity of magnetic disk devices.

If there is a media defect on the disk surface on which the positioning pattern is to be written, an erroneous position error signal is generated, which acts as a disturbance on the servo loop system. Furthermore, electrical noise that randomly enters the demodulation process of the position error signal also becomes noise in the position error signal. When the number of cells arranged in a limited servo unit is small, the influence of these disturbances is large and the positioning accuracy is degraded.

[0008] In order to obtain acceptable positioning accuracy,
A considerable number of cells are required, which necessitates making the servo section wider. This problem is more pronounced in the case of the sector servo system, which reduces the data area and prevents the device from increasing its capacity.

In the case of the tri-bit pattern, the cell length is the smallest among the patterns listed above and is desirable;
There was a problem caused by the asymmetry of the reproduced waveform. FIG. 11 is an explanatory diagram showing the asymmetry of the reproduction waveform of a single magnetization reversal. When reproducing the magnetization reversal recorded by the head, a symmetrical reproduced waveform 113 is supposed to be reproduced as shown by the broken line, but in reality it is an asymmetrical reproduced waveform 114 as shown by the solid line.
is obtained. The reason for this is due to the asymmetry of the magnetic field generated during recording and reproduction of the magnetic head. Furthermore, this tendency is even more pronounced in MIG (metal-in-gap) heads and thin film heads that are compatible with high-density recording and have been used in recent years. Also, due to the frequency characteristics of the filters and magnetic heads used in the reproduction circuit, the phase delay time differs depending on the frequency, so the waveform is distorted.
This becomes more noticeable as the recording density increases. Furthermore, since the waveform asymmetry depends on the electromagnetic conversion characteristics of the magnetic head and magnetic disk, the degree of asymmetry differs for each head, for each magnetic disk, and even for different locations on the same magnetic disk due to these variations.

FIG. 12 is an explanatory diagram showing an asymmetric reproduction waveform in a conventional tri-bit pattern, in which the positioning pattern is developed. The center lines 120 to 125 of each track during pattern writing are arranged at equal pitches I. The reproduced waveform 133 reproduced by the head 126 on the center line 122 of the track has a waveform with different wave heights as shown by the broken line in the figure. This is because when the magnetization reversal interval is short to a certain extent, adjacent asymmetric isolated reproduction waveforms (indicated by broken lines in the figure) with different directions of adjacent magnetization reversals interfere with each other, and the superimposed waveforms originally have the same wave height as described above. What should be makes a difference. A schematic representation of this waveform is shown at 134. Two positioning bits 136, 13
7 have different output amplitudes, but the head position error signal is calculated from the output amplitudes F and G of these positioning bits 136 and 137, and off-track (the amount of deviation between the track center and the head gap center) is proportional to F-G. do. Therefore, when F=G, the position error signal becomes 0, and the head is positioned at the center of the track. The center of the head in this case 129
is at a position shifted by H in the opposite direction every track with respect to the center line 122 at the time of pattern writing.

Similarly, the positions F=G of the reproduced waveform, that is, the respective track centers 127 to 132 in the reproduced waveform, are each shifted by the same amount from the center line at the time of writing. Therefore, the track pitch is O for odd-numbered tracks and E for even-numbered tracks. The relationship is O=I+2H and E=I-2H. As described above, the tri-bit pattern has a fatal drawback in that the track pitch is fundamentally different between even-numbered tracks and odd-numbered tracks. For this reason, if the head width was not sufficiently small relative to the track pitch, the head would protrude into the adjacent track and perform recording/reproduction, resulting in a problem that data on the adjacent track would be destroyed. Furthermore, since this pitch error depends on variations in electromagnetic conversion characteristics between the head and the disk, it differs between each head, each cylinder, and even within one track, and cannot be corrected using an algorithm or the like. In particular, in recent years when high recording density is required, MIG heads are
However, as films become thinner and linear recording densities and track densities continue to improve, it has been impossible to put tri-bit patterns, which have a fatal drawback of pitch error, into practical use.

There is no synchronization bit in the dibit pattern. First, the synchronization bit serves as a time reference for reproducing the positioning bit, and the reproducing circuit accurately samples the positioning bit while synchronizing with the synchronization bit to extract an accurate position error signal. In addition, a synchronization bit was essential in order to sufficiently track disturbance factors such as disc rotation fluctuations and time lags during pattern writing and extract position error signals. Furthermore, AGC (auto gain control)
In this operation, stable amplification and demodulation of the reproduced waveform is possible because there are synchronization bits having a constant output amplitude at regular intervals in the reproduced waveform. Therefore, dibit patterns without synchronization bits are susceptible to disturbance factors such as writing accuracy and have poor positioning accuracy, and have difficulty in stable amplification, resulting in poor S/N ratios, making them unsuitable for positioning in high-density magnetic disk drives. Met.

In the case of the burst method, although it depends on the configuration of the demodulation circuit, a long positioning pattern is required, making it difficult to increase the capacity of a magnetic disk device. Furthermore, it has been difficult to increase track density because positioning accuracy deteriorates because it is susceptible to the characteristics of the magnetic disk, such as local modulation and medium defects.

Therefore, the present invention is intended to solve the above-mentioned problems, and its purpose is to achieve high precision positioning by having a short cell length and being able to sample many position error signals within a limited area. In addition, the servo section is required to a minimum, the usable area for the data section is wide, and it is suitable for increasing capacity. Furthermore, it is a position with stable quality that is less susceptible to disturbances such as media defects, noise, and servo pattern writing errors. It is an object of the present invention to provide a magnetic disk device having a positioning pattern that can provide an error signal and achieve good positioning accuracy without pitch errors due to waveform asymmetry.

[0015]

[Means for Solving the Problems] A magnetic disk device of the present invention includes at least one magnetic disk, a spindle motor for mounting and rotating the magnetic disk, and recording and reproducing information on and from the magnetic disk. an actuator mechanism capable of moving the magnetic head to any position in the radial direction on the magnetic disk; a prime mover for driving the actuator mechanism; and the magnetic disk, the magnetic head, and the actuator. In a magnetic disk device having a container for storing a mechanism, (a
(a) the information recording surface of the disk has a plurality of tracks on which information is recorded concentrically; (c) the servo section has a positioning pattern made up of one or more groups; (d) the group consists of a plurality of cells; (e) the cells has a periodicity of n (n is an integer of 1 or more) track periods in the radial direction of the magnetic disk, (f) the cell has one or more synchronization bits and n positioning bits in the circumferential direction,
(g) the center of the radial recording width of the positioning bit is located approximately midway between the centers of adjacent tracks, and (h) the positions of the synchronization bit and positioning bit are geometrically located between adjacent tracks. (i) in each of said cells all synchronous bits have magnetization reversal; and (j)
Each track has magnetization reversal in the same direction in only one positioning bit within one cell, the circumferential position is different from each other in any track among n tracks, and (k) in the positioning bit of a certain track. The position of magnetization reversal in the circumferential direction is different for each cell in the same group, and (l) the presence or absence of magnetization reversal in the positioning bits is characterized in that the arrangement is the same in all groups.

[0016]

[Embodiment] An embodiment of the present invention will be described below. FIG. 1 shows a plan view of a magnetic disk device using a positioning pattern according to an embodiment of the present invention, and FIG. 2 similarly shows a sectional view. In the figure, a magnetic disk 1 is placed on a spindle motor 2 and is driven to rotate in a constant direction and at a constant speed. The magnetic head 3 floats above each surface of the magnetic disk 1 by the dynamic pressure of air, and performs recording and reproduction on the magnetic layer on the surface of the magnetic disk 1. The magnetic head 3 is attached to an actuator 4 and can be rotated to any position on the magnetic disk 1 in the radial direction. In this case, the actuator 4 is rotary and has a bearing 5 made of a ball bearing at the center of rotation. An actuator 4 is located at a position facing the magnetic head with respect to this bearing.
Voice coil motor 6 (hereinafter referred to as “VCM”) is the prime mover of
) will be placed. In this embodiment, the coil 7 of the VCM is fixed to the actuator 4 and the VCM including the magnet
The M magnetic circuit 8 is attached to a base frame 9, which will be described later. The magnetic flux generated in the magnetic circuit 8 of the VCM passes through the coil 7 of the VCM perpendicularly to the plane of FIG. can be given an arbitrary magnitude and angular acceleration in the direction of rotation. Spindle motor 2, actuator 4 and V
The magnetic circuit 8 of the CM is attached to a base frame 9, and forms a sealed space with a cover 10 to prevent dust from entering from the outside. A circuit board 11 is attached to the lower surface of the base frame. The circuit board 11 is equipped with electronic components such as the spindle motor 2, a drive circuit for the VCM 6, a recording/reproducing circuit, a servo circuit, a disk controller circuit, and the like.

FIG. 3 is a conceptual diagram of the recording surface of a disc in one embodiment of the present invention. All recording and reproduction is performed on concentric tracks 201. In this embodiment, each track is broadly divided into two areas in the circumferential direction. One is the data section 202, in which information that should originally be stored in the magnetic disk device is brought from outside the device and recorded. The other is a servo section 203 in which information for various controls of the magnetic disk device is recorded. The servo sections 203 are arranged at equal angular intervals in the circumferential direction of the track. The servo portion 203 has a circular arc having the center of rotation of the actuator as its center in the radial direction. In this embodiment, the arrangement of the servo section 203 and the data section 202 is the same on all disk surfaces.

FIG. 4 is a detailed view of the servo section 203 in one embodiment of the present invention. The arrows in the figure indicate the direction of magnetization. The servo part has the positioning pattern 20 described here.
In addition to 4, there is a synchronization part for synchronizing the reproduction circuit, a part to indicate the distinction of recording areas, and a track number (I
There are parts that represent D), but I will not discuss them in detail here. The positioning pattern 204 includes the actuator 4
A signal for positioning the track 201 on a predetermined track is written in advance, and a demodulation circuit, which will be described later, generates a position error signal from the center of the track 201 to be positioned. The positioning pattern 204 has a plurality of groups 205
Furthermore, each group has two cells 206.
It is more established. One synchronization bit S2 for each cell
21, and two positioning bits A222 and B223. As shown in FIG. 4, the center of the recording width of each bit is shifted by exactly 1/2 track from the center of the track. In reality, the recording width is determined by the width of the magnetic gap of the head, and is slightly narrower than the track pitch, but for simplicity, it is shown here as being the same. Sync bit S
There is always magnetization reversal in either positioning bit A222 or positioning bit B223. Therefore, in every cell, the direction of magnetization reversal of each of the synchronization bit and the positioning bit is the same, and the direction of magnetization reversal of the synchronization bit and the positioning bit is different. In this example, the arrangement of magnetization reversals in the positioning bits is repeated every two tracks. Further, the positions of the magnetization reversals of the positioning bits A222 and B223 in the circumferential direction are alternated every other cell.

The demodulation circuit determines the positioning bits A and B based on the synchronization bit 221 and the gate signal generated by the clock pulse obtained by the VCO oscillating in synchronization with the synchronization signal section in front of the synchronization bit 221. Samples the signal peak independently and holds the peak value. A position error signal is obtained by differentially amplifying these two held values.

FIG. 5 is an explanatory diagram showing the relationship between the positioning pattern of the magnetic disk device of the present invention, the reproduction waveform of the head, and the position error signal. The positioning pattern shown by the solid line in the figure is a part of the magnetization pattern (magnetization reversal) formed concentrically on the disk surface.For convenience, the fan angle is ignored and the radial direction is drawn by parallel vertical lines. It's drawn. Therefore, the disk rotation direction is indicated by the arrow 234, and the head 241 moves in a direction perpendicular to this. The head 241 is positioned at each track center line 235 to 240, and records and reproduces data. In the positioning pattern of this embodiment, the group 242 consists of two cells 224 and 225, and within each cell there are synchronization bits S221 and 243 having continuous magnetization reversal in each track, and a positioning bit A222 and a positioning bit B223 is arranged alternately for each track, and synchronization bit S243
They are arranged line-symmetrically. Such an arrangement is repeated every two tracks.

Next, this positioning pattern is used as the head 24.
The reproduced waveform demodulated by No. 1 will be explained at each position of the head 241. The head 241 moves to each position a, b, c, d, e between the track center line 237 and the track center line 238.
The reproduced waveforms when positioned at are a', b', c',
They are shown in d' and e', respectively. Taking reproduction waveform a' as an example, the head 241 is on the track center line 237, and the synchronization bit S221 and positioning bit A222 of the cell 224 are
, positioning bit B223, synchronization bit S of cell 225
243, positioning bit B223, positioning bit A2
They are played in the order of 22. Originally, if there is no asymmetry in the reproduced waveform, positioning bit A222 and positioning bit B22
3 have the same output peak value and have a reproduced waveform shown by a broken line 232 in the figure. However, due to the influence of the waveform asymmetry explained in the conventional example, the actual reproduced waveform is as shown in the solid line 233 in cell 2.
The output of positioning bit B of cell 24 decreases by D, and the output of positioning bit B of cell 22
The output of positioning bit A of No. 5 is also reduced by D. It is obvious that these two quantities are equal because the influence of waveform asymmetry acts equally on the two cells. Reproduction waveform b', c'
, d', and e', the positioning bit B of cell 224 is similarly
and positioning bit A of cell 225 cause an output drop by the same amount D.

Next, the process of generating a position error signal from the reproduced waveform will be explained. For convenience, position error signals are generated independently from cells 224 and 225, and are combined to generate a final position error signal. In the figure, the X axis indicates the magnitude of the position error signal, and the Y axis 245 indicates the position of the head 241 in the radial direction. For cell 224, the position error signal is

[0023]

[Math 1]

##EQU1## where K is a constant. If there is no waveform asymmetry, then at position a, that is, when the head is on the track center line 237, A and B are equal and X1 is 0. While the head moves to the center line 238 of the adjacent track, the position error signal becomes as shown by a broken line 226, and the track pitch P is constant between all tracks. If B is smaller than A by D due to waveform asymmetry, the position error signal will be the solid line 2.
27, and the track pitch is R. Also, this adjacent pitch is Q,

[0025]

[Math 2]

The relationship is as follows. Therefore, when the head moves in the radial direction Y1, the position error signal X1 of the cell 224 is
As shown by a solid line 229, the track pitch in the position error signal is repeated as R, Q, R, Q.

On the other hand, in the case of cell 225, positioning bit A
222, positioning bit B223 is synchronization bit S243
In contrast, the cell 224 is arranged at a symmetrical position. Therefore, the position error signal is

[0028]

[Math 3]

##EQU1## where K is a constant. If there is no waveform asymmetry, the position error signal will look like a broken line 226, similar to the cell 224, and the track pitch will be P, which is a constant value between each track. The actual reproduced waveform has asymmetry,
Due to the positional relationship of the positioning bits, A is smaller than B by D, contrary to cell 224. In this case, the position error signal becomes as shown by a solid line 228, and the track pitch becomes Q. Further, the pitch of this adjacent track is R. Therefore, the position error signal X2 of cell 225 indicates that the head is in the radial direction Y.
2, the track pitch in the position error signal is repeated as Q, R, Q, R, as shown by a solid line 230.

Finally, the position error signal is transmitted to cells 224,
It is generated by adding position error signals X1 and X2, which are obtained from the cell 225 and are influenced by waveform asymmetry. Therefore, the final position error signal is

[0031]

[Math 4]

It is expressed as: Waveform asymmetry effect D affects positioning bits A and B equally. Therefore, when added, the influence D of waveform asymmetry is completely canceled out and corrected. Naturally, the position error signal is also at a position parallel to the broken line 226 by the same amount as the solid line 227 and the solid line 228, so the final position signal This is equivalent to doubling, and the influence of waveform asymmetry is completely canceled out and the track pitch P is also constant between tracks. Therefore, when the head moves in the radial direction Y3, the final position error signal X3 becomes as shown by a solid line 231, where the track pitch P is constant and is not affected by waveform asymmetry at all. Since track pitch errors caused by waveform asymmetry can be completely eliminated in this way, the guard band (gap between tracks) does not need to be as wide as in conventional tri-bits, making it possible to increase track density. In addition, variations in the electromagnetic conversion characteristics of heads and disks did not affect pitch errors, and each head and cylinder achieved a good track pitch. Furthermore, since the cells 224 and 225 are placed very close to each other, variations in the original value of the position error signal due to output distribution on the disk, defects, and transient characteristics of the position error signal demodulation circuit are extremely small. Positioning accuracy within one track has also been significantly improved. Furthermore, MIG, thin film, and other heads compatible with high linear recording densities have become available for practical use for the above-mentioned reasons. The cell length of the positioning pattern according to the embodiment of the present invention is the same as that of the conventional tri-bit pattern and is the smallest among various patterns. Therefore, the number of samples for position error signal generation is large, and the influence of disturbances such as defects in the disk medium can be avoided. It has become smaller, more reliable, and the positioning accuracy during recording and playback has been further improved.

On the other hand, in the sector servo system, since the number of servo parts can be reduced, the data part can be widely utilized, contributing to further increase in capacity.

Furthermore, since the influence of waveform asymmetry caused by the synchronization bits can be completely corrected, the positioning bits can be sampled at accurate times while synchronizing with the synchronization bits, which increases the accuracy and reliability of the position error signal. Disturbance factors such as disc rotational fluctuations and circumferential (temporal) fluctuations during pattern writing can be eliminated. Furthermore, stable AGC operation enabled demodulation of a position error signal with a high S/N ratio, making it possible to perform excellent positioning.

The embodiment described above is an example of the minimum configuration in which the repetition period of the positioning bit is 2 tracks and the number of cells in one group is 2, but the positioning in a more general embodiment of the present invention is A detailed diagram of the pattern for use is shown in FIG. Here, only the positioning pattern 204 of the servo section is shown. The positioning pattern 250 is divided into several groups 205, and further groups 20
5 is composed of four cells 206 here. Within each cell the positioning bits are 251, 252, 25
There are four locations, No. 3 and 254, and magnetization reversal occurs at any one location. Furthermore, the positioning bits are arranged differently in all four cells. With this arrangement, as in the previous embodiment, the four bits are equally located immediately before the synchronization bit 221 and are easily affected by waveform interference, so that the position error signal due to waveform asymmetry is reduced. Changes are averaged out.

In the embodiments described above, the position error signal obtained by the demodulation circuit was one, that is, a one-phase signal, but the details of the positioning pattern according to another embodiment of the present invention are shown in FIG. In the figure, the positioning pattern 20
4, group 205 has its positioning bit 222,
The center of the recording width of the positioning bits 223 is shifted from the center of the track 201 by 1/2 of the track pitch, and the center of the recording width of the positioning bits 248 and 249 of the group 207 coincides with the center of the track 201. Therefore, by demodulating each group individually, the phase difference of each group is 90%.
It is possible to extract two-phase position error signals that are shifted by degrees. Of the two-phase position error signals, the one with superior linearity is used sequentially for each half track, allowing accurate control no matter where the actuator is located.

In another embodiment, the track number (
By applying this method to, for example, a gray code portion representing a track number, there is little variation in the radial position at which the track number changes, and the length of that portion in the circumferential direction can be made relatively small. If this signal is also used as one of the two-phase position error signals, a greater effect can be obtained.

Furthermore, positioning pattern 2 according to the present invention
By providing the servo section 203 including 04 on one entire surface of the disk 1, position error signals can be obtained with high temporal density (that is, the band is increased), which enables more accurate positioning of the actuator and faster speed. Movement becomes possible. In this case as well, a position error signal with higher density and more accurate synchronization can be obtained than in the conventional method.

[0039]

As described above, according to the present invention, by equalizing the order of positioning bits and synchronization bits in each cell, the time axis of the head, disk, and recording/reproducing circuit system can be adjusted in all tracks. The influence of directional asymmetry can be minimized, and the accuracy of the position error signal can be greatly improved, which in turn improves the control accuracy of the actuator.

[0040] Furthermore, compared to the case where similar accuracy is achieved using other positioning signal recording methods, the time length of the positioning pattern can be shorter, and therefore the length of the data section can be made larger. This greatly contributes to increasing the capacity of devices. Furthermore, according to the present invention, by averaging and equalizing the positioning bits as much as possible, disturbances in the position error signal due to local output fluctuations (modulation, media defects) and fluctuations in recording/reproducing characteristics that are destined for magnetic disks can be avoided. It is possible to make it smaller. Therefore, the control characteristics of the device can be maintained at a constant level, greatly contributing to quality stability.

[Brief explanation of the drawing]

FIG. 1 is a plan view of a magnetic disk device using a positioning pattern according to an embodiment of the present invention.

FIG. 2 is a sectional view of a magnetic disk device using a positioning pattern according to an embodiment of the present invention.

FIG. 3 is a conceptual diagram of a disk recording surface in an embodiment of the present invention.

FIG. 4 is a detailed diagram of a servo section in an embodiment of the present invention.

FIG. 5 is an explanatory diagram showing the relationship between a reproduction signal waveform and a positioning signal with respect to a positioning pattern of a magnetic disk device and a position of a magnetic head in an embodiment of the present invention.

FIG. 6 is a detailed view of a positioning pattern in another embodiment of the present invention.

FIG. 7 is a detailed diagram of a two-phase signal positioning pattern in still another embodiment of the present invention.

FIG. 8 is a detailed view of a positioning pattern showing cells of a conventional dibit pattern.

FIG. 9 is a detailed diagram of a positioning pattern showing cells of a conventional tri-bit pattern.

FIG. 10 is a detailed view of a positioning pattern showing cells of a conventional modified dibit pattern.

FIG. 11 is an explanatory diagram showing the asymmetry of reproduced waveforms.

FIG. 12 is an explanatory diagram showing an asymmetric reproduction waveform in a conventional tri-bit pattern.

[Explanation of symbols]

1 Magnetic disk 2 Spindle motor 3, 106, 241 Magnetic head 4 Actuator 6 Voice coil motor 7
VCM coil 8 VCM magnetic circuit 9 Base frame 11 Circuit board 111, 112, 221, 243 Synchronization bits 136, 137, 222, 223, 250 Positioning bit 201 Track 202 Data section 203 Servo section 204 Positioning pattern 205, 207, 242 Groups 206, 224, 225 cells

Claims (6)

[Claims] 1. At least one magnetic disk, a spindle motor on which the magnetic disk is mounted and rotationally driven, at least one magnetic head for recording and reproducing information on and from the magnetic disk, and the magnetic disk. A magnetic disk device comprising: an actuator mechanism capable of moving a head to any position in the radial direction on the magnetic disk; a prime mover for driving the actuator mechanism; and a container housing the magnetic disk, the magnetic head, and the actuator mechanism. (a) the information recording surface of the disk has a plurality of tracks on which information is recorded concentrically, and (b) at least one of the information recording surfaces of the disk has a servo used to control the actuator. (c) the servo section has a positioning pattern composed of one or more groups; (d) the group consists of a plurality of cells; (e) ) the cell has a periodicity of n (n is an integer of 1 or more) track periods in the radial direction of the magnetic disk; (f) the cell has one or more synchronization bits and n positioning bits in the circumferential direction; (g) the center of the radial recording width of the positioning bit is located approximately midway between the centers of adjacent tracks, and (h) the positions of the synchronization bit and the positioning bit are located between adjacent tracks. geometrically continuous, (i) all synchronous bits in each of said cells have magnetization reversals, and (j) each track has magnetization reversals in the same direction in only one positioning bit within one cell; (k) The circumferential position of the magnetization reversal in the positioning bit of a certain track is different for each cell in the same group, and (l) ) A magnetic disk device characterized in that the presence or absence of magnetization reversal in the positioning bits is arranged in the same manner in all groups. 2. The magnetic disk device according to claim 1, wherein the number of cells included in the group is n. 3. The magnetic disk device according to claim 1, wherein the synchronization bit and the positioning bit have equal time intervals. 4. The magnetic disk device according to claim 1, wherein in at least one group, the center of the radial recording width of the positioning bit is approximately equal to the center of each track. 5. The magnetic disk device according to claim 1, wherein the entire recording surface of at least one of the magnetic disks is used for a servo section. 6. The magnetic disk device according to claim 1, wherein data sections and servo sections are arranged alternately in each track, and the servo sections are arranged at approximately equal intervals in the circumferential direction.