JP2000156051A - Magnetic disk device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique for minimally suppressing an increase of an alternate sector memory and an alternate processing time at a defect adjacent register time and improving recording data reliability by alternating adjacent sectors of the smallest number before a defect occurring later on a magnetic disk grows. SOLUTION: The read data are saved beforehand into a nonvolatile memory or a magnetic disk for an error being a recoverable read error and recovering by offset read (step 104). Thereafter, an error relevant sector is rewrited with a data pattern suited to defect detection (step 105), and then, the error relevant sector is rereaded, and the reproducibility of the error is confirmed (step 106). At this time, only when the recovery error by the offset read is reproduced (step 107), the alternate processing is executed for the releavant sector and the adjacent sector in the direction opposite to the offset recovery (step 108).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an increased number of replacement sector memories by replacement of a minimum number of adjacent sectors before a subsequent defect of a magnetic disk grows in a magnetic disk drive after shipment from a factory. In addition, the present invention relates to a technique for minimizing a replacement processing time at the time of registration of a defect adjacent to improve the data reliability of a magnetic disk device.

[0002]

2. Description of the Related Art Defect registration of a magnetic disk is performed strictly at the time of manufacturing, and registration of an expected sector adjacent to a defective sector is also performed. However, for a defect occurring later at the user, only the error sector is registered as a defect. Also,
It was impossible for the user to replace a track adjacent to the error sector as preventive maintenance.

[0003]

Since the recording density of a recent magnetic disk has rapidly increased and the track interval has become narrower, defects on the magnetic disk due to dust, growth of protrusions on the magnetic disk, and the like, have been reduced. It also extends to adjacent tracks. Therefore, in the related art, when the replacement command (reassign command) or the automatic replacement setting is performed, only the error-related sector is replaced. However, as the recording density is further increased, the error-related sector and its neighboring sectors are changed. Should be replaced to improve data assurance.

An object of the present invention is to replace a minimum number of adjacent sectors before a subsequent defect grows on a magnetic disk, thereby minimizing the increase in replacement sector memory and the replacement processing time at the time of defect adjacent registration. An object of the present invention is to provide a technique for improving the reliability of recorded data.

[0005]

SUMMARY OF THE INVENTION According to the present invention, a recoverable read error is determined to be necessary for adjacent defect registration for an error recovered by offset read, and the sector and the recovered offset direction are determined in the opposite direction. Register adjacent defects. When this method is used, it is possible to perform adjacent registration with a minimum number, so that an increase in the replacement sector memory can be suppressed to a minimum. In addition, at the time of adjacent registration, the replacement processing time can be reduced.

[0006]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a flowchart showing an example of registration of an adjacent defect, FIG. 2 is a block diagram showing an example of the configuration of a magnetic disk drive, and FIGS. 3 and 4 are examples of a defect on a magnetic disk. is there.

First, a method for detecting a defect in a magnetic disk and a method for registering an adjacent defect shown in FIG. 1 will be described with reference to FIGS.

The magnetic disk drive 20 according to the present embodiment
HDA21 (head disk assembly) and P
It consists of CB22 (printed circuit board).

The HDA 21 includes a plurality of magnetic disks 23 on which servo information and data are written, and a spindle motor 24 mounted with the magnetic disks 23 and rotated.
A plurality of heads 26 for recording and reproducing information on and from a recording surface of the magnetic disk 23, and a head 2 via an arm 27.
6 and the R / W integrated circuit element 25 (R / W
A carriage 28 that carries a WIC) and moves the head 26 in the radial direction on the magnetic disk 23 to perform a positioning operation, and a voice coil motor 29 (VCM) that drives the carriage 28.

The PCB 22 is connected to a spindle motor driver 30 for controlling the rotation of the spindle motor 24, a servo control circuit 31 for controlling the position of the head 26 in the circumferential direction of the magnetic disk 23, and to the R / WIC 25 of the HDA 21 for recording data. R / W control circuit 32 for performing reproduction, R / W
The MPU 33 instructs the W, servo, and spindle rotations collectively and transmits and receives commands to and from the host. The data error detection 35, the defect adjacent determination 37, and the defect registration processing unit 38 are performed in the MPU 33, and when it is determined that there is a defect on the magnetic disk 23, the defect information is recorded in the defect memory 34.

FIG. 3 and FIG. 4 show an example of a late defect of a magnetic disk. In the case of FIG. 3, the error sector 43 of the N track becomes a recovery error by offset reading in the N-1 track direction. Therefore, the defect exists in the N + 1 track direction. In this case, the sector 43 and the sector 44 on the N + 1 track in the direction opposite to the recovered offset direction are registered as defects. In the case of FIG. 4, the recovery error recovers without the offset read. Therefore, only the sector 52 is registered as a defect. By determining the defect registration location in this way, the number of adjacent defect registrations is minimized, and the increase in defect registration memory is reduced.

In the case of the present embodiment, a recoverable read error that has been recovered by offset read is registered as a defect adjacent.

As an example of defective adjacent registration, a defect of a recoverable read error is determined by the method shown in FIG. 1, and a defective sector replacement process is performed. The process proceeds to the defect adjacent registration process only when a replacement command is issued from the HOST for the recoverable read error (step 101) or when the automatic replacement process is performed (step 102). If recovery is performed by offset read, it is determined to be a defect adjacent registration target (step 103). If recovery is performed in other cases, replacement processing is performed only for the sector (step 109).

In the case of defective adjacent registration, first, read data is saved to a nonvolatile memory or a magnetic disk in order to guarantee replacement source data (step 104). Thereafter, a data pattern suitable for defect detection is written in the sector in which the error has occurred (step 105), thereby improving the defect detection rate and eliminating the influence of offset writing due to disturbance during data recording. Next, read is performed again to confirm the reproducibility of the error (step 106). Here, when the recovery error due to the offset read is reproduced (Step 10)
Only in step 7), a replacement process is executed for the sector and the adjacent sector in the direction opposite to the offset recovery (step 10).
8).

In the present embodiment, the defect adjacent processing time excluding the data assurance processing due to unexpected power shutdown is the read retry processing time + write time + reread retry processing time + replacement seek time + replacement. This is the destination data write time. When the read retry processing time is 3 rotations and the average seek time to the replacement destination is 1 rotation, the above time is 3 rotations + 1 rotation + 3 rotations + 1 rotation + 1 rotations in total of 9 rotations.
Short rotation and defect adjacent processing time.

The selection of the presence or absence of the defect adjacent registration is made by HOST.
More arbitrary command can be specified.

Next, the unconditional adjacent registration logic will be described. The defect adjacent registration is performed without using the defect adjacent registration and without determining logic the number of adjacent tracks specified in both directions of the error sector. The selection of the presence or absence of the unconditional adjacent registration logic and the setting of the number of defective adjacent tracks can be arbitrarily specified by a command from a higher order. When unconditional adjacent registration is performed for the defect in FIG.
, The sector 43, the sector 42 on the (N-1) th track, and the sector 44 on the (N + 1) th track are replaced. The other features of the present invention are listed below.

(Feature 1) For recoverable read error,
A magnetic disk drive comprising adjacent registration logic for registering a defect in the direction opposite to the recovered offset direction with respect to an error recovered by an offset read.

(Feature 2) For recoverable read error,
A magnetic disk drive comprising an unconditional adjacent registration logic for registering a defect on both sides without a decision logic.

(Feature 3) For recoverable read error,
A magnetic disk device which executes adjacent registration logic or unconditional adjacent registration logic when a replacement command is received or when automatic replacement is set.

(Feature 4) A magnetic disk drive having a function of selecting adjacent registration logic and unconditional adjacent registration logic, and arbitrarily setting the number of adjacent tracks of unconditional adjacent registration by a command from a host device. .

When the number of defective adjacent tracks is designated as 2,
The sector 43 and the sector 41 on the N-2 track,
A total of 5 sectors, that is, the sector 42 on the N-1 track, the sector 44 on the N + 1 track, and the sector 45 on the N + 2 track are replaced. The number of defective adjacent sectors for unconditional adjacent registration is
Assuming that the number of defective adjacent tracks is M, the number is 2M + 1. Therefore, when M is increased, it is necessary to increase the size of the defect information memory.

Thus, the user can arbitrarily set three types of defect registration methods. In the first method, only the relevant sector is replaced as in the past, and the prevention of the late defect is not performed. Second
The method uses a neighbor registration logic to register a minimum number of defective neighbors,
Implement optimal late defect prevention. The third method uses unconditional neighbor registration logic to more severely replace the growth of subsequent defects.

[0025]

According to the present invention, the necessity of registration of an adjacent defect is determined, and the adjacent registration can be performed with a minimum number only when necessary. The effect that the replacement processing time can be shortened is obtained.

[Brief description of the drawings]

FIG. 1 is a flowchart illustrating an example of adjacent defect registration according to the present invention.

FIG. 2 is a block diagram illustrating an example of a configuration of a magnetic disk drive according to an embodiment of the present invention.

FIG. 3 is a cross-sectional view illustrating an example of a defect extending between tracks on a magnetic disk.

FIG. 4 is a cross-sectional view showing an example of a minute defect on a magnetic disk.

[Explanation of symbols]

Reference numeral 20: magnetic disk drive, 21: HDA (head disk assembly), 22: PCB (printed circuit board),
23 ... magnetic disk, 24 ... spindle motor,
25 ... R / W integrated circuit device (R / WIC), 26 ...
Head, 27 ... arm, 28 ... carriage, 2
9: Voice coil motor (VCM), 30: Spindle motor driver, 31: Servo control circuit, 32 ...
R / W control circuit, 33: MPU, 34: defect information memory, 35: error detection unit, 36: defect registration determination unit, 37: defect registration unit, 40: defect, 41
... Sector on N-2 track, 42 ... Sector on N-1 track, 43 ... Sector on N track,
44 ... Sector on N + 1 track, 45 ... Sector on N + 2 track, 50 ... Defect, 51 ... Sector on N-1 track, 52 ... Sector on N track, 53 ...
Sector on track N + 1.

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G11B 20/18 572 G11B 20/18 572F

Claims (1)

[Claims] 1. A magnetic disk drive, wherein a minimum number of adjacent registrations is performed before a defect of a magnetic disk becomes large with respect to a recoverable read error.