JP2696429B2 - Two reference track techniques - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information storage system.

2. Background Art Mass storage for computers and other information systems is commonly provided by magnetic media storage systems, such as rigid or flexible disk storage. A rotating disk having a layer of magnetic media on the surface is accessed by a read / write head used to store information on and retrieve information from the surface of the disk. In order to store information on a magnetic media disk, a magnetic flux reversal is induced in the magnetic particles constituting the surface. As the read / write head passes through this flux reversal, a signal is generated at the head and the signal can be decoded to obtain information.

Data is stored in a series of concentric "tracks" on the surface of the magnetic disk. The read / write head moves back and forth over the radius of the disk and can be selectively positioned on one of the tracks. After being positioned on the track, the track rotates under the head remaining in place, allowing the head to write and read data on the track.

For effective writing and reading, it is necessary to know the position of the track with respect to the head. In addition to knowing on which track the head is located, it is necessary to know where on the track the head is located. Conventionally, position information is obtained by using a servo pattern. The servo pattern is a permanent pattern written on the storage disk and can be used to obtain position information. The servo pattern can be detected by the servo head and represents the track position by proper decoding. The servo pattern is also written as a series of concentric tracks. In some multi-disk storage devices, all of one side of the storage disk is dedicated for servo information. The servo disk is accessed by a servo head and the position information is read. Since the servo head has a fixed positional relationship with respect to the read / write head, the position of the servo head can be used as an indication of the position of the read / write head. In addition to having a dedicated surface for servo information, a "sector" servo pattern can be used. The sector
In a servo pattern, the pie wedge of the servo pattern is sandwiched between sections of data information.

For efficient data storage in a magnetic disk system, it is desirable to maximize the data storage capacity. All of the disk surface dedicated to servo tracks cannot be used as data tracks. One way to create free space for data tracks is to embed a servo layer below the surface of the data disk itself. One such technique is the applicant's US patent application 07 / 116,109, entitled "Servo Pattern".
It is described in. In that application, a plurality of servo lines extending from an inner track to an outer track are formed on a data disk. These servo lines are used as part of an embedded servo layer, and the surface of the disk is dedicated to data storage. The servo pattern is written on each side of the disc, and if the disc is transparent,
When viewed from above, each servo line appears to intersect. Servo heads located on each side of the disk detect the intersection of servo lines. By comparing the time differences at the intersection of the corresponding tracks on each side of the disk, the radial position of the head is determined and the servo and data tracks are defined.

Typically, a disk drive assembly uses separate data and servo heads. In one example, the data head and the servo head are part of a dual core assembly, where one core of the dual core assembly is a data head.
The other core acts as a servo head. In order to read from a disk written by another disk drive assembly, the relationship between the data head and the servo head must be known. For example,
Assuming that the distance between the data head and the servo head of one dual-core assembly is one track wider than the other dual-core assembly, the disk written by one drive is replaced by the other drive. , An error of one track occurs. Since the spacing between the servo tracks is on the order of hundreds of micro inches, any small change in the spacing between the data head and the servo head will result in an error of several track widths.

Such changes are typically detected and compensated for at power-up of the drive assembly. An adaptive calibration procedure is performed to compare the relative positions of the servo head and the data head with the stored nominal values. For example, when the data head is located on the first data track, the servo head is located on the servo track "n" if the distance between the servo head and the data head is a nominal value. I have. If not at the nominal value, the servo head
It is located at track “n ± m”. Here, m is the number of tracks of the offset caused by a change in the interval between the servo head and the data head.

This adaptive calibration technique is effective only when the absolute position of the servo track is known.
Such a request implies the use of periodic missing bits in the servo pattern. In order to reliably detect a pulse in a waveform including a missing bit, peak detection by a differentiating network is required. This places severe restrictions on the type of filter used to remove incoming noise from the data head during its writing. Servo patterns that do not require peak detection are often desired, in which case the absolute position of the servo track is unknown.

Calibrate the drive and adjust the servo
To determine the distance between the head and the data head,
The conventional adaptive calibration technique described above cannot be used. If you are puzzled to use a disc written on one drive for another drive,
The above intervals must be known. There are usually manufacturing tolerances in the spacing between the data head and the servo head in a dual head assembly.

In many conventional disk drives that use servo tracks for track tracking in a closed loop, the servo track spacing is even and equal to the data track spacing. In such drives, data track #
It is only necessary to perform positioning using 0 as a reference point. In such a drive, data track # 0 serves as a reference track and only one reference point is needed.

In some conventional servo techniques, as described in the applicant's pending U.S. patent application Ser. No. 07 / 116,109, Servo Patterns, the servo tracks are spaced close together. First, the distance (in inches) between servo tracks decreases continuously as one moves from the outer circumference to the inner circumference of the disk. Second, the position of each servo track is defined only in the time domain. As a result, interpolation between servo tracks is extremely easy. Only a clock higher than the frequency of the servo pulse is needed.

Some servo tracks using unequally spaced servo tracks store the servo track coordinates of each data track in a ROM (read only memory). In that case, the table in the ROM is valid only for the head whose interval between the two heads is the nominal value for all the drives to be processed. If not, a non-linear correction factor must be used to calculate the actual servo track corresponding to each data track on the drive.

Accordingly, it is an object of the present invention to provide a method and apparatus for detecting a deviation from the nominal data head and servo head spacing.

It is another object of the present invention to provide a method and apparatus for calibrating a disk drive assembly such that a disk written on one drive can be accessed on another drive. .

Yet another object of the present invention is to provide a "missing bit"
Discs without the need for servo patterns
It is to provide a method and apparatus for calibrating a drive assembly.

It is yet another object of the present invention to provide a method and apparatus for calibrating a disk drive assembly without requiring peak detection of servo patterns.

SUMMARY OF THE INVENTION The present invention is a method and apparatus for performing adjustments of a disk drive assembly and determining, from a nominal value, an offset value that compensates for a servo head difference with respect to a data head position. The present invention can be applied to a relative track counting servo system in which the absolute value of a track cannot be known from reading of the servo track itself. In such a technique, the anchor track must be determined to determine the position of the data head, and the servo track information must be determined relative to the anchor track. In the preferred embodiment, a single head assembly having both a data head and a servo head is used.
The distance between the center of the data head and the center of the servo head is
It can be seen that a disc written by one drive assembly can be read by another drive assembly. This distance is a nominal value, but may change during the manufacturing process.

In the present invention, both data and servo heads are used to determine the servo anchor track and to determine an offset correction value that corrects for changes in servo / data head distance. For example, as an example, assume that the head is within plus or minus 10 micro inches. Once the data head is in place, the servo track below and near the servo head is determined to be the servo anchor track. The head is then moved across the disk until the data head is centered on the second reference track. During this movement, the number of servo tracks between the first reference track and the second reference track is counted, including the minute parts on both sides. This number is compared to a table that stores the number of tracks when matching the data / servo head distance to a nominal value. The difference between the actual number and the stored value is
Used to generate an offset value (in inches) that corrects for changes in data / servo head distance during seek and other operating positions. In this embodiment, two reference tracks must be at known positions. The two reference track techniques of the present invention can implement a single frequency servo pattern. This enables a single zero-crossing detection in the servo detection circuit and eliminates the need for peak detection.

The arrangement of the servo tracks in this embodiment is very reproducible. A tool, known as a servo writer, is placed on each disk's servo pattern and two reference tracks.
Servo patterns are very precise (plus or minus)
10 micro inches). The two reference tracks are written with a radius whose absolute value is not known and only the spacing is known. If the spacing between the servo head and the data head is known to the servo writer, the exact radius of each reference track is known. The distance between the two reference tracks must be controlled and known. Assume that a disk written with a servo writer as described above has been placed in a drive in which the spacing of the data head relative to the servo head is smaller than that of the servo writer.
The data head of the drive is located outside the reference track. Then, the servo head of the drive is arranged with a radius larger than the radius of the servo writer. This corresponds to the fact that the servo tracks of the drive are at a coarser interval than the servo writer. The data head moves inward to the reference track while counting the servo tracks. The servo head is again at a greater radial distance from the center than the servo writer. This also corresponds to the fact that the servo tracks of the drive are coarser than the servo writers. This is because the number of servo tracks when the data head moves from the outer reference track to the inner reference track is smaller than when the data head moves with the servo writer. From the difference in the number of servo tracks, the difference between the arrangement intervals of the data head and the servo head between the drive and the servo writer can be calculated.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a plan view of a servo disk according to a preferred embodiment of the present invention.

FIG. 2 is a plan view of the data / servo dual core.

FIG. 3 is an enlarged partial view of the disk for explaining the operation of the present invention.

FIG. 4 is a block diagram of a read circuit according to the present invention.

(Detailed Description of the Invention) Perform the adjustment of the disk drive assembly,
A method and apparatus for determining an offset value that corrects a difference of a servo head with respect to a position of a data head from a nominal value is described. The detailed figures in the following description, the nominal value of the distance between the servo head and the data head, the number of servo tracks between the reference tracks, etc., are provided to completely describe the present invention. However, those skilled in the art can practice the invention without such numbers. Accordingly, well-known parts have not been described in detail so as not to obscure the present invention.

The present invention advantageously uses a dual core. The present invention is applied to the case where the difference from the nominal value between the servo head and the data head is known, especially when the absolute position information of the servo track is not known. The present invention has been described only in connection with the servo technique described in US patent application Ser. No. 07 / 116,109. However, the present invention is not limited to only such servo pattern techniques, but can be used for many servo-based techniques.

In this embodiment, the data storage disk has a plurality of first servo lines embedded below a data layer on one surface of the disk. A plurality of second servo lines are embedded under the second data layer on the other side of the disk. This servo
The line is formed by a logarithmic spiral extending from the inner peripheral side to the outer peripheral side of the disk. The servo lines gather toward the inner circumference. By making the servo line a logarithmic spiral, it intersects the servo line at a certain angle with a radial line of the disk. The servo head at the end of the actuator arm assembly, which moves in the radial direction of the disk, servos at the same angle at each point on the servo line.
Intersect with the line. As a result, the maximum signal can be detected by the servo head at all radial positions.

Since the servo lines are arranged at an angle on the concentric data tracks, crosstalk between the data and the servo tracks is minimized. As the angle between the read head and the information track increases, the amount of signal detection decreases until the azimuth angle, at which point the minimum detection occurs. Maximizing the angle between the servo lines and the data tracks maximizes the azimuth effect, minimizes undesirable obstructions, and reduces crosstalk between the two tracks. The present invention uses a time reference technique in which the servo tracks are independent of the data tracks. In this way, a relatively large servo head can be used and the embedded servo can be easily detected.

In a preferred embodiment of the present invention, the adaptive calibration technique can be used with flexible magnetic media. However, those skilled in the art will appreciate that the present invention is equally applicable to magnetic tape and other rotating storage media such as Winchester hard disk drives, magneto-optical and even optical drives.

A plan view of the disk used in the preferred embodiment of the present invention is shown in FIG. Figure 1 shows the outer circumference 13 and the inner circumference
It is a storage disk with 14. The first reference track 11 is
A second reference track made near the outer circumference 13 of the disk 10
The 12 is made near the inner circumference 14 of the disk 10. The first and second reference tracks 11, 12 are read by the data head. A plurality of servo tracks 15 are formed on the disk. In a preferred embodiment of the present invention, data tracks are written on the surface of disk 10 and servo tracks are formed as part of the buried layer technique in which the tracks are written in the surface of disk 10. Multiple data tracks
It can be formed between the first and second reference tracks 11, 12 for storing and retrieving information. The access and positioning of these data tracks are controlled by the servo track 15. When determining the position of a servo track on a servo track, the corresponding data track can be identified.

The appearance of the dual core data / servo head is
It is shown in FIG. The dual core head 27 includes a servo head 28 and a data head 29. The servo·
The head 28 has a gap 25 and a dual head 29
Has a gap 26. Servo gap 25 makes an angle of approximately 45 ° with data gap 26. As described above, crosstalk between the data layer and the servo layer is reduced due to the azimuth effect. In the structure of FIG. 2, the length of the servo gap 25 is about four times the length of the data gap. In prior art servo technology, the width of the servo head gap is approximately equal to the width of the data head gap because the servo tracks coincide with the data tracks. In the use case of the present invention where the servo pattern is independent of the track pitch, the size of the servo gap width is independent of the size of the data gap width. This allows the use of relatively large servo gaps and facilitates the detection of buried layers in the preferred embodiment of the present invention. Further, a head having a large gap width is less expensive to produce than a head having a small gap width, and reduces the cost of a servo head having the structure of FIG.

The servo head 25 and the data head 26 are separated by an interval 17. The nominal value of the spacing in the preferred embodiment is about 60 mils. This number is only applied in this case, and other numbers are also considered in the present invention. If the absolute position of the servo track is unknown, it is necessary to recognize the deviation of the distance from the nominal value in order to allow movement of the disk from drive to drive.

Since the servo pattern time in the example of the present invention is used as a reference, it is necessary for the servo head to detect the servo signal of each data track. The time-based technique of the present invention is a presence / absence technique that allows the use of large servo signals and has an improved signal-to-noise ratio over conventional servo techniques.

Although one dual core head 1 is shown in FIG. 2, in any case where there is a fixed nominal relationship between the servo head and the data head,
The present invention is equally applicable, and the deviation in the above relationship is:
It must be determined for the storage and retrieval of information. The invention has a special use case where the absolute servo position is not available from the servo track. Further, in the illustrated dual core technique, the data head and the servo head are at an angle to each other. This relationship is not a requirement of the present invention, but serves only as an application. Dual core heads, in which the servo head and the data head are in a corner, are also contemplated by the present invention.

Referring again to FIG. 1, the first and second reference tracks 11, 12
Is written to the disk 10 at a known position according to the coordinates of the servo track, relative to absolute inches. That is, the first reference track is always written on the servo track, and the second reference track is written on the servo track n + 5. In practice, the data head is located on the center line of one of the reference tracks, for example the first reference track 11. At that time,
The servo tracks between the servo heads are defined as servo track 0 or anchor servo tracks. The data head is then moved to the second reference track 12
Move to a position above. While this movement is taking place,
The number of servo tracks traversed during the movement is counted.
This count is compared to a stored number representing the number of servo tracks when the centerline 25 of the servo head gap and the centerline 26 of the data head gap are equal to the nominal distance.

If the number of servo tracks counted has changed from the stored number, an offset correction signal is issued to compensate for the deviation from the nominal value of the distance. The basic equation for determining the position in the radial direction in a pattern having a logarithmic spatial spread is as follows.

Equation 1 R _T = R _Z θ or R _T = R _Z θ ^-KN _T where R _T = Radial position in inches of a given servo track R _Z = Position of servo track 0 N _S = ^{Flux of} servo head Number of passes per rotation of 2 α = angle of servo head K = (π / N _S ) (cotα) N _t = number of tracks of a given servo track π = 3.1415926... The relative difference between the servo tracks (n ₁ and n ₂ ) is denoted as ΔS, and R ₁ > R ₂ Let ΔS _{1 be} the nominal difference between the positions of the two reference tracks in the relative servo track, and read ΔS ₂ in the relative servo track under given drive conditions Is the nominal difference between the positions of the two reference tracks. A space between the data head and servo head, when the difference between the nominal value and delta ² S, Equation _{^{3 Δ 2 S = ΔS 1 -ΔS}} 2 Here, ε represents the difference between the space between the data head and the servo head and the nominal value.

Therefore, a change in the number of tracks corresponds to a change in the space between the data head and the servo head. By utilizing this relationship, it is possible to compensate for spatial variations.

Table 1 is obtained from Equation 3 with R ₁ = 1.54482 inches and R ₂ = 0.88043 inches. This is ε = −0.
The representative value of Δ ² S in the range of 003 inches to ε = + 0.003 inches is shown.

Table 1 ε (inch) Δ ² S (servo track) −0.003 −6.062 −0.002 −4.037 −0.001 −2.017 0 0 0.001 2.013 0.002 4.023 0.003 6.029 The operation of the present invention is shown in FIG. In FIG.
The space of the servo track is not included. In the preferred embodiment, the present invention, the space is logarithmic. However, it appears that the application of the present invention is being made in other non-regular servo pattern fields. Whenever the drive power is increased or calibration is required, Figure 3
The data head is located on the outer reference track 11 as shown in FIG. On the reference track 11, a series of bursts are written as alternate bursts. The magnitudes of the left and right signals of the read channel of the data head are captured and compared. When the left and right sizes match, the data head is located on the center line of the reference track 11.

When the data head is located at the center line of the reference track 11, the servo head is accessed. When the data head is at the above position, the servo track 15A below the servo head is defined as servo track 0. The relative servo track counter is reset to zero in this case.

Next, the data head reads the second reference track 12 of 19B.
Move to the center line. During this “seek” operation,
The relative servo track counter is incremented for each servo track. When the data head is located on the center line of the second reference track, the count of the track counter is captured and represented as ΔS ₂ in Equation 3. The nominal value of the track counting is ΔS _1, Δ ² S is readily determined.

If the space between the data head and the servo head is less than or equal to the nominal value, as shown at 20A, then the number of crossed servo tracks will be less than if the space is at the nominal value. Like 20B,
If the space between the data head and the servo head is greater than or equal to the nominal value, the number of tracks that intersect while the data head moves from the first reference track to the second reference track will be greater than or equal to the nominal value.

The arrangement and centering of the data head on the center line of the first and second reference tracks can be performed using a track following circuit of the data head as a seek engine. According to FIG. 1, radial position information 16 is written on the disc for use by the data head.
This position field has three zones for providing relative radial position information to the data head during positioning. The data head moves to the centerline of the reference track in a two step process. In a first coarse stage, the data head reads information from the position field to determine its relative position with respect to the reference track. Section 16A of the position field contains a pattern indicating that the data head is outside reference track 11. Section 16B contains a pattern indicating that the data head is between the reference tracks. section
16C contains information that the data head is further inside the inner reference track 12. The data head is
This coarse relative position information is used for focusing on the reference track. In the preferred embodiment, this coarse step has a resolution of plus or minus 1000 microinches.

After the data head is over the reference track, the data
The amplitude comparison technique can be used as a fine mode in which the head is brought over the center line of the reference track. The amplitudes of the signal bursts on the left and right channels are sampled and adjusted until the head position matches the amplitude.
The fine positioning step has a resolution of plus or minus 10 micro inches in the preferred embodiment of the present invention.

After the offset is determined, the correction factor is
Used to guarantee changes in data head spacing. In this embodiment, the value Δ ² S is used to calculate the correction value. For example, when a seek command is given to the drive, a look-up table is used to direct the servo head to the servo track corresponding to the desired data track. This look-up table is based on the nominal data / servo interval. Therefore, the value Δ ² S
Is used to correct the number of servo tracks (corresponding to the desired number of data tracks) by an amount that depends on how much the offset between the data and the servo head differs from the nominal spacing.

FIG. 4 is a block diagram illustrating the readout circuit of the present invention. The dual-core data / servo head 27 of the present invention outputs a data signal 31 and a servo signal 32 to the counter 33, respectively. Head assembly 27 is first
If it is above the reference track, the counter 33 is reset to zero. As head 27 moves to the second reference track, the counter is incremented by each servo track traversed during that movement. Counter 33 outputs on line 29 a signal indicating the number of servo tracks between the first and second reference tracks. The output is compared to a value representing the number of tracks when the servo / data head spacing is at a nominal value.
This value is stored in the lookup ROM 34 and supplied to the input of the comparator 36. The output of the comparator 36 is supplied to seek control means 35 such as a built-in microprocessor. The output 36 of the comparator 36 is used as an offset value for correcting a seek command based on the difference in interval. seek·
When a command 40 is input to the seek control means 35, the seek control means accesses the ROM 34 via line 38 to determine which servo track to seek. Signal 42 is used to correct this value so that the seek corrects for changes in servo / data head spacing. The output 41 of the seek control means 35 is a signal applied to the actuator to move the head 27 to the correct servo track.

Thus, a method for determining and adjusting for changes in data / servo head spacing has been described.

 ──────────────────────────────────────────────────の Continued on the front page (72) Inventor Visejack, John United States 95120 San Jose Porto Eilgra, California 5896 5896

Claims (2)

(57) [Claims] A servo system having a fixed positional relationship with each other.
In a method of carrying a head and a data head, forming a plurality of servo tracks on a storage medium, forming a first reference track on the storage medium, and forming a second reference track on the storage medium. Forming a track; determining, when the data head is positioned on the first reference track, an anchor track as a servo track below the first servo head; Counting servo tracks by counting the number of servo tracks traversed by the servo head when moving from one reference track to the second reference track; and a nominal value between the servo head and the data head. And compares the stored track number, which represents the number of servo tracks, and the servo track count value. The steps that the servo is the difference that is proportional to the difference away from the nominal value of the distance between the servo head and data head
Generating a first error signal by taking the difference between the track count and the stored number; and a method for calibrating a servo head and a data head comprising: 2. A servo system having a fixed positional relationship with each other.
A system for storing on a rotating medium having a plurality of servo tracks formed on a surface of a head and a data head and a rotating storage medium, wherein a distance between the data head and the servo head is calculated from a nominal value. Forming a first reference track on the surface; forming a second reference track on the surface; placing a data head on the first reference track; Moving the head from the first reference track to the second reference track; counting the servo tracks to obtain a servo track count when the data head moves; A servo track count value from the stored track number representing the number of servo tracks when the distance from the head is a nominal value; Subtracting; and utilizing the difference to determine the actual distance difference between the servo head and the data head. How to determine the difference from the nominal value.