JPH10269729A - Disk unit - Google Patents

Abstract

(57) [Summary] [PROBLEMS] In an ID-less format method, track identification can be performed without recording track identification information in a servo, a servo WR process time can be reduced, and sector identification can be verified. To improve reliability. SOLUTION: The head is a dual head of an RD head and a WR head, and for the purpose of reading only at the time of data WR, cylinder number information, head number information, servo sector address information, etc. are provided at offset positions where the RD head passes at the time of data WR. Is recorded at the time of data WR, the servo sector address information 39 read from the servo sector, the designated cylinder number information 42 and the head number information 43, and the SCF are read from the SCF. The calculated cylinder number information, head number information, and servo sector address information 41 are compared by a comparator 49, and a track is verified by an SCF error determination circuit 47 based on the comparison result.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk drive for improving the reliability of an ID-less format system and reducing costs by shortening the manufacturing time.

[0002]

2. Description of the Related Art In recent years, computers have been improved in performance and miniaturization, and there has been a strong demand for data storage devices, particularly magnetic disk devices, to have a smaller size, a larger capacity and higher performance. In response to this demand, attention has been paid to a method of increasing the capacity by improving the format efficiency. As an example of this approach, in a conventional disk device, there is a method of deleting an ID area provided for identifying a sector as described in Japanese Patent Application Laid-Open No. 5-174498.
(Here, this method is defined as "ID-less method").

FIGS. 2B and 2C show an example of a disk format using this method. Hereinafter, the configuration of this format will be described. The disk format includes a servo sector (SSCT16) provided for controlling a position on a recording medium of a head for performing a data recording / reproducing operation, and a data sector (DSCT17) provided for storing user data. It is roughly divided. As shown in the SSCT example of FIG. 2B, the servo sectors include an AGCG (automatic gain control gap) 18, an SVMK (servo mark) 1
9, IDXMK (index mark) 20 / SCTM
K (sector mark) 21, TRK (track number) 2
2, a POS (position signal) 23 and an ISG (intersector gap) 24.

[0004] The AGCG 18 is an area provided for adjusting a read gain of servo information, and the SVMK 19 is an area provided for detecting a head position of a servo sector. I
DXMK20 / SCTMK22 is an area provided for identifying the head of a track or sector.
K22 is an area for storing a full track number. The POS 23 stores detailed head positioning information between tracks, and controls each of a detailed positioning operation (settling) and a follow-up operation (following) that always positions the head on a target track. used. The ISG 24 is an area provided for absorbing fluctuations in the rotation of the disk.

On the other hand, the data sector 17 shown in FIG.
In the case of adopting a format in which the recording density is constant over the inner and outer circumferences of the disk, the configuration differs between the DSCTB format in which the servo sector is included therein and the DSCTA format in which the servo sector is not included. DSC
TA is ISG24, PLO (frequency synchronization pattern) 2
5, BS (byte sink) 26, DATA27, ECC
28 and a PAD 29. The ISG 24 is an area provided for absorbing fluctuations in the rotation of the disk as described above. The PLO 25 is an area provided for synchronizing clocks with read data.
S28 is an area provided for detecting a timing of changing from serial data to parallel data (for example, byte data). The DATA 27 is an area for storing user data, and the ECC 28 is an area provided for correcting an error in the read-out DATA 27 when the error is found.

[0006] The DSCTB basically has a format in which the SSCT 16 is inserted into the DSCTA, but the PLO 25 and the BS 26 are arranged again after the SSCT 16. This is because at the time of reading, when passing through the SSCT 16, the read processing is interrupted once, so that it is necessary to perform clock synchronization and byte synchronization again to perform the read processing again.

In this method, since the data sector 17 does not have an ID portion storing the data sector address (DSA), the data sector address (DSA) is set to the servo sector 1
6 using the servo sector address (SSA).
The SSA sets the address of the servo sector 16 including the index mark 20 to 0, and sets the subsequent servo sector 1
Each time the sector mark 21 of No. 6 is detected, the address can be obtained by incrementing the address by one. S
If the SA is obtained, the address conversion table is used to determine the SS.
A DSA corresponding to A can be obtained. This address conversion table can be easily created from the ratio of the number of servo sectors 16 and the number of data sectors 17 arranged in a track in units of a track.

[0008]

In the IDless format system, the address of the data sector 17 is detected by the following method.
Index mark 2 stored in servo sector 16
This is performed by converting a servo sector address recognized by detecting and counting 0, a sector mark 21, and a servo mark 19 indicating a head position of the servo sector 16 into a data sector address using a conversion table or the like. These are the identification of data sectors on the track where the head is located.In a disk device having a plurality of recording surfaces and a plurality of heads, a track different from a desired track is erroneously selected due to a failure or malfunction of the head switching circuit. In such a case, the erroneous selection cannot be detected, resulting in erroneous recognition of the data sector. There is a possibility that WR is performed on the erroneous data sector to destroy user data, or erroneous data is transferred to the host computer. For this reason, it is required to detect the track where the head is actually located by some method.

In the prior art, as shown in the SSCT example, the full track number is stored in the servo sector 16 and the servo sector 1 in which the head is actually positioned is located.
6 shows a method of directly reading the full track number in No. 6 and identifying the track. When the full track number is recorded in the servo sector, the servo WR by the servo WR device needs to be performed for each track since the number of all the servo sectors is different.

In a disk drive having a plurality of recording surfaces for recording cylinder numbers in servo sectors and a plurality of heads,
In the same cylinder, even if the recording surface is different, the number is the same, so that a plurality of heads can perform servo WR at the same time. For example, in the case of a disk drive having a plurality of n recording surfaces and a plurality of n heads, the servo W
R can be performed. In this case, the servo WR of one cylinder can be performed by one rotation of the disk, and the servo WR of all the tracks may be the servo WR operation for the number of cylinders.
That is, it is possible to perform the WR of the servo of all the tracks in the disk rotation time for the number of cylinders.

However, in the method of recording a full track number, in the case of a disk device having a plurality of recording surfaces and a plurality of heads, a plurality of heads cannot perform servo WR at the same time.
Servo WR is performed for each head. That is,
In order to perform servo WR for all tracks, a rotation time for the number of tracks (the number of cylinders × the number of recording surfaces) is required.

In general, servo recording is performed using a dedicated servo WR device in a servo WR process at the time of manufacturing a disk device. Therefore, the amount of servo WR that can be performed at one time is limited, and the method of recording a full track number requires much time in the servo WR process as compared with the method of recording a cylinder number.

The problem of offset in a disk device employing a dual head will be described with reference to FIG. FIG. 3 is a conceptual diagram of a sector of a disk device of a general ID format. At the time of DATA WR, the WR head is positioned on the center line A of the track to perform WR, and the DATA section records the data with the center line A of the track as the center. At this time, the RD head passes on the B line due to the offset of the RD head / WR head. DATA
At the time of RD, RD is performed by positioning the RD head on the center line A of the track. This is because when reading is performed on the recorded position A line, the read signal level is large and the read error is small. ID section is W when DATA RD
In order to read out at the time of R, recording is generally performed by positioning the WR head on the center line C of the A line and the B line. However, DATA
In both RD / WR cases, the RD head is
Since the read is shifted from the center line C of D, the read error rate may be deteriorated. If a reading error occurs in the ID section, there is a possibility that ID cannot be reliably identified,
By performing the retry of reading the ID again, the throughput is deteriorated. Further, since the data is read out of the center line C of the ID, the data is easily affected by the neighboring tracks, and the distance between the tracks cannot be narrowed.

An object of the present invention is to improve the reliability of the above-mentioned ID-less format system by using a means for performing track identification without recording track identification information on the servo while improving the reliability of the servo WR process time of disk drive manufacturing. Is to improve the reliability by means of shortening the time and verifying the sector identification.

[0015]

In order to solve the above-mentioned problems, the present invention provides a head for recording / reproducing data, a plurality of data blocks provided on a recording surface for storing data, and the head. A recording medium having a recording surface including a servo block for storing control information for controlling a position on the recording surface, and a servo check field for storing positioning information for determining a positioning error. A device for storing positioning information in the servo check field; a unit for reading out the positioning information only at the time of data write access;
The apparatus includes: means for comparing the read positioning information and the position information read from the servo block with a target position of a data write to determine a positioning error; and a means for controlling writing of a data block according to a positioning error determination result. Like that.

The track offset between the data block and the servo check field provided for determining the positioning error is determined by the RD head and the WR.
It is set to match the offset amount of the head. In addition, in the case of a read error of the servo check field, a control means for referring to a read history of at least one servo check field before, and when there is a non-error read in the read history, a control means for allowing a read error of the servo check field. Is provided. Further, a means is provided for separately controlling the read start timing of the servo check field and the write start timing of the data block. Further, all of the above-mentioned means are accommodated in one LSI.

[0017]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below. Each servo sector, as shown in FIG.
GCG, SVVMK, IDXMK / SCTMK, TRK,
It consists of POS and ISG. AGCG is an area provided for adjusting a read gain of servo information, and SVMK is an area provided for detecting a head position of a servo sector. IDXMK / SCTMK is an area provided for identifying the head of a track or sector, and TRK is an area for storing a cylinder number. In the present invention, track verification is performed using the SCF described later.
In particular, it is not necessary to record track identification information such as a full track number in a servo sector. Even in a disk device having a plurality of recording surfaces and a plurality of heads, a cylinder number is recorded in this embodiment. The POS stores detailed positioning information between cylinders, and is used for controlling each of a detailed positioning operation (settling) and a following operation for always positioning the head on the target cylinder (following). ISG is an area provided for absorbing fluctuations in disk rotation.

This servo sector is generally recorded using a dedicated servo WR device. This is because, in order to position the head on the track, it is necessary to perform head position control using, for example, a dedicated positioning device using a laser. In the present invention, the servo WR method is not a problem and is a generally used method.
R may be performed. In this embodiment, since the contents of the servo sector are common in the cylinder, servo WR is possible for a plurality of heads at the same time. For this reason, it is desirable that the servo WR device is set so that the signal processing device can be set to the servo WR mode so that all the heads can simultaneously perform WR, and the servo is simultaneously recorded for each cylinder for each cylinder. The description of the data sector is omitted because it is not different from the above-mentioned conventional example.

The operation of each block will be described with reference to FIGS. 1 and 5 by taking data write operation in the ID-less format system as an example. The CPU 13 controls the entire disk processing device.

The host IF control unit 10 manages data transfer from the host computer 15 to the buffer control unit 9. It also receives commands from the host computer and sends them to the CPU. The CPU 13 analyzes the command, calculates the address of the data sector for storing the data on the recording medium surface, and obtains the TGTCYLREG. (Target cylinder register) 42, TGTHDREG. (Target head register) 43, TGTSCREG. (Target sector register) 44 is set in each register. The buffer control unit 9 performs control of temporarily storing data from the host IF control unit 10 in the data buffer 14 and transferring the data to the drive IF control unit 8. The ECC control unit 6 performs an EC operation for detecting / correcting a read error in a read process.
Generate C. The drive IF control unit 8 manages data transfer and performs sector processing. The servo control unit 12 reads a servo sector for each servo sector,
1 and the head is set to TGTCYLREG. Position on the track designated by. Also, the servo control unit 12
Output IDXPLS indicating the starting point of the track, SCTPLS indicating the boundary of the sector as shown in FIG. 2A, and SRVPLS indicating the servo area. The signal processing device 5 is T
GTHDREG. 43, and switches the data from the drive IF control unit to the R / W amplifier 4
Transfer to The transferred data is recorded on the data surface recording medium via the R / W head.

SSAREG. (Servo sector address register) 39 is SSAREG. Controlled by the update control circuit 40, reset is performed by IDXPLS, incremented by SRVPLS, and the servo sector address is stored. The data sector address generation circuit 45 generates the SSARE
G. FIG. The data sector address where the R / W head is located is obtained by 39 and SCTPLS. Sector control circuit 4
6 and the data sector address obtained by the data sector address generation circuit 45 and TGTSCREG. 44
To determine whether it is a desired data sector. If it is determined that the data sector is a desired data sector, the drive IF sequencer 36 turns on the SCTMATCH signal. The drive IF sequencer 36 is started by the CPU after head switching and track positioning are completed, and controls the drive IF control unit 8 to execute data WR processing on a data sector in the case of SCFOK described later. SCF described later
If an error occurs, the data WR processing is not performed,
An error report is sent to the CPU 13.

The SCF (servo check field) which is a feature of the present invention is recorded by the data processing device after the servo WR by the servo WR device. In this embodiment, as shown in FIG. 2A, the SCF 30 is arranged at every other servo sector and behind the servo sector. This is an example of the embodiment, and an arbitrary number of SCFs may be arranged at an arbitrary position.
The SCF avoids the aforementioned dual head offset problem by aiming to read only at the time of DATA WR. That is, when recording the SCF,
Based on the servo sector information, the WR head is positioned at the offset position where the RD head passes at the time of DATAWR, that is, on the B line in FIG. 4, and the SCF is recorded. Therefore, at the time of DATA WR, the WR head is positioned on the line A in FIG. 4 and the RD head is positioned on the B line in FIG.
Since the RD head reads on the center line B of the SCF, the above-described problem of the offset of the dual head is avoided. The servo sector is recorded over the entire width of the track.

As shown in FIG. 2D, each SCF 30 includes ISG, PLO, BS, CYL, HD, SSA, E
It is composed of CC and GAP. The ISG 24 is an area provided for absorbing fluctuations in the rotation of the disk.
LO is an area provided for synchronizing clocks with read data. The BS 26 is an area provided for detecting a timing of changing from serial data to parallel data. CYL31 is an area for storing a cylinder number, and HD32 is an area for storing a head number. It is checked whether the track is a desired track by using CYL and HD. SSA (servo sector address) 33 is SC
The address of the servo sector before F30 is stored.
The sector address recognition operation is checked using the SA.

The ECC 35 reads the read CYL, HD, S
This area is provided for determining whether there is an error in the SA and correcting the error.

The information recorded in the SCF is an example of the embodiment, and various information may be recorded. For example, only the head number is recorded and S is used only for verification of head switching.
CF may be used. Further, information indicating a track state, information for setting a write-protected track of identification information such as a recording track of system information or a defective track, or the like may be recorded. An error check code such as a CRC may be added for the purpose of error check, instead of an ECC for performing error correction. Also, without taking the configuration as in this embodiment, it is possible to read without PLO,
The track identification information may be recorded at a sufficiently low frequency.

FIGS. 1 and 7 show the SCF processing operation.
The SCF control sequence of FIG. 6 and an example of SCF determination operation timing of the SCF control sequencer of FIG. 6 will be described.
The SCF control sequencer 37 is a drive IF sequencer 3
6 is started by the CPU 13 before starting the data WR operation (701). After startup, the system waits for SCFPLS (702). SCFPLS is a SCFPLS generation circuit 3
As shown in FIG. 2A and FIG. 6, the data is output for each position of the SCF 30. In this embodiment, since the SCF is arranged every other servo sector, the SCFPLS generation circuit 38 generates the SSAREG. When the least significant bit of 39 is 0, it is output after the servo sector. SSAREG. 39 is IDX
It is reset by PLS and incremented by SRVPLS to indicate the servo sector address, which serves as a reference for data sector address recognition.

After the SCFPLS is detected, the BS is searched as shown by BSSRCH in FIG. 6 to perform the SCF read operation (703). If the BS cannot be detected within the specified time, the SCF is determined to be indeterminate and the next SCF is determined.
Wait for CFPLS. When a BS is detected within a specified time as shown by BSF in FIG. 6 (see BSF in FIG. 6)
(704), read CYL, HD, SSA (CCYLTEN, CHDLTEN, CSSALTEN in FIG. 6)
Reference, where LT means latch, EN means enable), CCYLREG. , CHDREG. ,
CSSAREG. Each current SCF capture RE
G. FIG. 41 (705, 706, 707). 70
If the time is over in the BS detection of No. 4, the SCF is determined to be indefinite (715).

When an ECC error is detected (708)
Performs an error correction operation (709). In the case of an uncorrectable error (see ECC error determination and correction in FIG. 6) (7)
10) makes this SCF indeterminate SCF (71)
6). If there is no ECC error (708) or correction is possible (710), the current SCF capture REG41
CCYLREG. , CHDREG. , CSSARE
G. FIG. And the contents of TGTCYLREG. 42, TGTH
DREG. 43, SSAREG. A comparison is made for each of the contents 39 (see SCFCHKEN in FIG. 6) (71).
1). TGTCYLREG. 42 and TGTHDRE
G. FIG. 43 to verify that the track is the desired track. By comparing with 39, S
SAREG. 39 and SSAREG. The update control circuit 40 is verified. If the comparison result is a match (712), the SC
FOK (713), and if they do not match (712), S
CFNG is set (714). SCF uncertain, SC
The results of FOK and SCFNG are stored in the SCF result latch shown in FIG. 8 in the SCF error determination circuit 47 for each SCF process (see SCFCHKLTEN in FIG. 6) (71).
7).

In this embodiment, the SCF on the track
Inability to detect BS due to the occurrence of medium defects in the area, EC
Means are provided for preventing the throughput from deteriorating due to the retry process and the occurrence of sectors in which data WR becomes impossible when an uncorrectable read error occurs. According to this means, when the SCF is uncertain, the SCF error is not immediately determined, and the previously read SCF check result is referred to.
If this is SCFOK, SCF is not determined and SC
Judge as FOK. In this embodiment, the configuration is such that the SCF up to two SCFs is referred to. The SCF result latch has the current SC as the shift register configuration of the number of SCFs to be referenced as shown in FIG.
The F determination result and the previous SCF determination result are held. In the present embodiment, a three-stage shift register configuration is used as shown in FIG.

Referring to FIG. 8, SC in the means of this embodiment will be described.
The F result determination logic will be described. The SCF latch 1 stores the recently read SCF result, and the SCF latch 2
Stores the immediately preceding SCF result, and the SCF latch 3 stores the immediately preceding SCF result. These are updated every SCF processing. In the determination of the SCF1 result (801),
In the case of SCFNG, it is determined that an SCF error has occurred.
Is determined as SCFOK, and the SCF latch 1
If it is determined to be indeterminate, the process proceeds to the determination of the SCF2 result (802) without immediately determining that an SCF error has occurred.

In the determination (802) of the SCF2 result,
In the case of SCFNG, it is determined that an SCF error has occurred.
Is determined as SCFOK, and the SCF latch 2
If it is determined to be indeterminate, the process proceeds to the determination of the SCF3 result (803) without immediately determining the SCF error.

In the determination of the SCF2 result (803),
In the case of SCFNG, it is determined that an SCF error has occurred.
Is determined as SCFOK, and the SCF latch 3
If it is determined to be indeterminate, it is determined to be an SCF error.

That is, when the SCF latches 1, 2 and 3 are all undefined, or when the SCF latch 1
, The SCF latch 1 is indeterminate and the SCF latch 2
In the case of CFNG, SCF latches 1 and 2 are undefined and SCF
If the latch 3 is SCFNG, it is determined that an SCF error has occurred.

The above-described SCF determination processing is performed independently of the drive IF sequencer. When an SCF error occurs, an SCF error signal is turned on for the drive IF sequencer. For this reason, the SCF read start timing and the data block write start timing are separately controlled.

[0035]

As described above, according to the present invention, I
In a disk device adopting the D-less format system, by providing a track and sector verification means, it is possible to prevent user data from being destroyed due to a failure or malfunction of the head switching circuit, and to improve the reliability of the device and the system.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a configuration of a drive IF control unit according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating an example of a disk format and a relationship between the disk format and various pulses.

FIG. 3 is a diagram showing an example of a sector configuration in an ID format format employing a dual head.

FIG. 4 is a diagram showing an example of the arrangement of SCFs when a dual head according to an embodiment of the present invention is employed.

FIG. 5 is a diagram showing a system configuration of a disk device.

FIG. 6 is a diagram illustrating an example of an SCF control operation timing;

FIG. 7 is a diagram showing an example of an SCF control sequence.

FIG. 8 is a diagram illustrating an example of SCF error determination.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Disk device 2 Data surface recording medium 3 R / W head 4 R / W amplifier 5 Signal processing device 6 ECC control unit 7 Data processing device 8 Drive IF control unit 9 Buffer control unit 10 Host I / F control unit 11 Motor driver 12 Servo control unit 13 CPU 14 Data buffer 15 Host computer 16 SSCT 17 DSCT 30 Servo check field SCF 36 Drive IF sequencer 37 SCF control sequencer 38 SCFPLS generation circuit 39 SSAREG. 40 SSAREG. Update control circuit 41 Current SCF capture REG. 42 Target CYLREG. 43 Target HDREG. 44 target SCTREG. 45 data sector address generation circuit 46 sector control circuit 47 SCF error determination circuit 48 data transfer control circuit 49 comparator

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Akira Nakama 2880 Kozu, Odawara-shi, Kanagawa Prefecture Storage Systems Division, Hitachi, Ltd.

Claims (5)

[Claims] 1. A head for recording / reproducing data, a plurality of data blocks provided on a recording surface for storing data, and control information for controlling a position of the head on the recording surface. What is claimed is: 1. A disk device comprising: a recording medium having a recording surface including a servo block for storing and a servo check field for storing positioning information for determining a positioning error, wherein positioning information is stored in the servo check field. Means for reading the positioning information only at the time of data write access; means for comparing the read positioning information and the position information read from the servo block with a target position of data writing to determine a positioning error; Means for controlling writing of a data block according to a positioning error determination result. Disk apparatus. 2. The disk device according to claim 1, wherein a track offset amount between the data block and a servo check field provided for determining the positioning error is adjusted to an offset amount between an RD head and a WR head. A disk device characterized by the above-mentioned. 3. The disk device according to claim 1, wherein at the time of a read error of the servo check field, a read history of at least one servo check field is referred to, and when the read history includes a non-error read, A disk device comprising control means for allowing an error in reading the servo check field. 4. The disk drive according to claim 1, further comprising means for separately controlling a timing to start reading a servo check field and a timing to start writing a data block. 5. The disk device according to claim 1, wherein all of the units are contained in one LSI.