JP3533309B2 - Servo control method and magnetic disk drive - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a servo control method and a magnetic disk device for enabling use of a medium having a defective servo information section and avoiding write failure due to an abnormal position signal.

[0002]

2. Description of the Related Art As a servo control method in a recent magnetic disk device, an embedded servo method in which a servo information section is embedded between data sections as shown in FIG. 1 is mainly used. The servo information section is written in several hundred thousand places on the entire disc surface and occupies 3 to 4% of the entire surface. There are hundreds of medium defects on the entire surface that affect the servo information.
Dozens of places may collide with the servo information section.

FIG. 2 is an enlarged view of the servo information section. AM
It is divided into (address mark) part, gray code part, and burst part. The AM part is a part where the reference timing for gray code and burst acquisition is recorded, and the gray code records physical position information on the disc. The burst is a portion used for generating a position signal indicating a deviation amount (off-track amount) from the on-track position. When a medium defect hits each point, A
An M-miss, a gray code defect, and a position signal abnormality due to a burst abnormality occur, which causes a seek error.

An HDA (head disk assembly) on which servo information has been written by a servo track writer (equipment for writing servo information on a disk) is subjected to a seek operation test by a servo information tester, and seek error (AM). Whether or not the device can be used is determined based on the presence or absence of an AM miss due to an abnormality, a gray code failure due to a gray code bit abnormality, and a position signal output abnormality due to a burst portion abnormality. In the past, HDA in which seek error occurred
All had a disc replacement regeneration process as defective. However, with regard to AM misses, a seek error occurs if a plurality of consecutive occurrences occur, but if there is only one place, the servo information in the relevant servo information section is not used, and a rescue measure is performed by referring to the immediately preceding servo information. It was taken.

[0005]

Recently, MR (MAGNETO RESISTIVE) heads have begun to be used in magnetic disk devices, and the recording density has rapidly increased and the track spacing has become narrower. Therefore, even a small defect that has not affected the servo information in the past has reached a level that cannot be ignored when an abnormality occurs in the servo information. Therefore, the rate of failure during the seek operation test and the product inspection test is high, and the yield of HDA is low. Also, M
TA (THERMAL ASPERITY) peculiar to R head and TA
Subsequent defects also occur due to traces, film peeling, and the like.

When a defect occurs in the burst portion of servo information, the position signal becomes abnormal and writing becomes impossible due to a large amount of misalignment (hereinafter, write fault). Depending on the degree of the defect, a reproducible write fault occurs.
Even the retry operation in which the write operation is repeated again cannot be repaired, resulting in an unrecoverable error. Further, if an unrecoverable error occurs during use by the user, the system downs and the magnetic disk device is replaced as an important failure of the magnetic disk device. An object of the present invention is to improve the yield of HDA, and an object of the present invention is to relieve an unrecoverable error of a write fault during use by a user as a recoverable error in a retry operation.

[0007]

In order to achieve the above object, the present invention provides a servo control method in a magnetic disk device for generating a position signal based on servo information written on a medium surface and performing positioning control. That is, each servo information section on the medium is sequentially read, and it is determined whether or not there is an abnormality in the servo information of the read servo information section. When it is determined that there is an abnormality, one of the servo information sections with the abnormality is I try to erase the information in the department.

Further, it is a servo control method in a magnetic disk device for generating a position signal based on the servo information written on the medium surface to perform positioning control. When a write fault occurs during the data writing operation,
Check the location of the error on the medium surface, register the location of the error in the registration memory, write the data again, and when the write fault occurs again, check the location of the error on the medium surface. When the error occurrence location and the contents of the registration memory are compared, and the comparison result is a match,
Sampling processing is not performed on the servo information portion at the error occurrence location.

Further, in a magnetic disk device that generates a position signal based on servo information written on the medium surface and performs positioning control, there is an abnormality in the servo information of each servo information portion on the medium that is sequentially read. There is provided a means for determining whether or not there is an abnormality, and a means for erasing a part of the information of the servo information portion having the abnormality when it is determined that there is an abnormality.

Further, in a magnetic disk device that generates a position signal based on servo information written on the medium surface and performs positioning control, when a write fault occurs during a data write operation, an error occurs on the medium surface. Confirmation means for confirming the location, means for registering the location of the error confirmed by the confirmation means in the registration memory, and instructing the data write again, and when a write fault occurs again in the data write again When the comparison result by the comparison means for comparing the error occurrence location on the medium surface confirmed by the confirmation means with the contents of the registration memory and the comparison result by the comparison means are the same, the servo information part of the error occurrence location is A means for stopping the sampling process is provided.

[0011]

FIG. 3 shows the configuration of an embodiment of a magnetic disk device to which the present invention is applied. A magnetic disk device includes a plurality of discs on which servo information and data are written, a spindle motor that mounts the discs and rotates, a plurality of heads that record and reproduce information, an R / W IC and a head that are fixed. Carriage for moving on board, VCM
HDA (head disk assembly) consisting of (voice coil motor), spindle motor driver 1, V
A CM driver 2, an R / W channel 3, a drive control circuit 7 including an R / W control circuit 4, a servo control circuit 5 and a servo microcomputer 6, and a SCSI control circuit 8 for controlling a command from a host controller are mounted. It consists of a PCB (package board). Servo information is recorded on the entire disk surface of the HDA in advance by a servo track writer as shown in FIG. The servo information is reproduced by the head and read as an output signal R
/ W IC and then input to R / W channel 3. The servo information shaped by the R / W channel 3 is
It is input to the servo control circuit 7. The servo data detector 9 in the servo control circuit 7 detects the position information (AM portion, gray code) and outputs it to the servo microcomputer 6, and the peak hold circuit 10 detects the position deviation information and A / D converts it. Is output to the servo microcomputer 6. The servo microcomputer 6 controls the VCM driver 2 on the basis of the position information and the position shift information to perform the seek operation of the servo and the following control. Also, a write fault determination during the write operation is performed based on the position signal based on the positional deviation information.

The "AM crushing remedy" which is an embodiment for improving the yield of HDA of the present invention will be described below. The AM crush relief is performed at the HDA manufacturing stage. A
The M crush remedy utilizes the fact that in the remedy of the conventional technique, when an AM miss occurs, the servo information in the servo information section is not used and the control is performed with reference to the servo information in the immediately preceding servo information section. That is, the AM crush relief is to crush, that is, erase, the AM portion of the servo information portion where the defect has occurred, thereby causing an AM miss. The AM crush is the servo microcomputer 6 when it is determined that the servo information section read by the servo microcomputer 6 is abnormal based on the position information (AM section, gray code) and the positional deviation information input from the servo control circuit 5. Is performed by generating an erase signal, outputting the erase signal to the R / W control circuit 4, and under the control of the R / W control circuit 4, the AM portion of the servo information portion in which the abnormality has occurred is crushed. .

FIG. 4 shows a conceptual diagram of the AM crushing process.
AM is counted up from the time of occurrence of the servo sector pulse created based on the servo information of the normal servo information section.
When the timing of the AM portion of the crushing target servo sector is reached, the servo microcomputer 6 outputs an erase signal to the R / W control circuit portion 4. Based on this erase signal, the R / W control circuit section 4 stops the data and performs the write operation to erase the AM section.

FIG. 5 shows an example of the servo information portion which has been subjected to the AM crushing process. This example shows a case where there is a medium defect in the burst part of the servo information part.
Since the M part is erased and the AM part cannot be detected,
The position information based on the gray code and the position shift information based on the burst signal in the servo information section cannot be acquired, and the information in the servo information section immediately before normal is used.

The defective servo information portion to be crushed by AM is determined by the error content of the test result at the manufacturing stage and executed by manual intervention. It is also possible to automatically perform the test during the test by setting the judgment condition of the defective sector in the program of the servo microcomputer.

Next, an embodiment of "servo information unnecessary relief" for relieving an unrecoverable error due to a write fault when used by a user will be described below. The above-mentioned “AM crush relief” can be applied at the manufacturing stage, but it is not a process that can be executed during use by the user after shipping. Even after shipment, the T.P. Servo information failure due to A occurs.

FIG. 6 shows an abnormal position signal due to a defect in the burst portion. The level of the position signal changes for each servo information section. In the case of FIG. 6, there is one defective servo information, but the position signal is abnormal for several places due to the transient response. During write operation due to abnormal position signal, the off-track write inhibit threshold is exceeded and a write fault occurs. No matter how many times the write operation is repeated in this state (usually, the magnetic disk device performs a retry operation for an error), a write fault occurs, and if the number of retries is exceeded, a write failure unrecoverable error occurs and the magnetic disk device fails. Will be. "Servo information unnecessary relief" is used to check and store the error occurrence location when a write fault occurs during data write operation, and when a write fault occurs again during retry operation, the error occurrence at this time occurs. Check the location and determine if there is a servo information failure at the same location as the stored error occurrence location. If a servo information failure occurs at the same location, the defective servo information section By stopping the sampling process for (1) and ignoring the defective servo information, the position signal abnormality is prevented from occurring and the write operation is normally terminated.

FIG. 8 shows a detailed sequence of the servo information unnecessary relief processing. When the data write operation starts (801), it is determined whether or not a write fault has occurred in the data write operation (802). If no write fault has occurred, the process ends normally, and if a write fault occurs, the retry mode is entered and the retry mode is set ( 803), and the number of retries N is set to 0 (804). The upper limit of the number of retries is set to M in advance. Next, the value of the registered memory (memory for setting an error sector in which the servo information section is abnormal) is reset (805). Next, the number of retries N is checked (806), and when it is determined that the number of retries N is equal to or greater than the set number M, the retry mode is reset (807) and an abnormal end (unrecoverable error) is reported. . If it is determined that the number of retries N has not reached the set number M, the data write operation is performed again (808), and 1 is added to the number of retries N (809). Then, determine the occurrence of a write fault (8
10) If it is determined that the write fault has not occurred, the retry mode is set (811), and a normal end (recoverable error) is reported. If it is determined that the write fault has occurred, the location of the error is confirmed (812). It is determined whether the confirmed error occurrence location is the same as the value set in the registration memory (813). If it is determined that they are not the same, the error occurrence location is set in the registration memory ( 81
4) Retry. That is, the process returns to the check step (806) of the number of retries N. If it is determined that the error occurrence point is the same as the value in the registered memory, the servo information disuse process (815) for stopping the sampling of the servo information at the error occurrence point is performed, and the retry is performed.

FIG. 7 shows a position signal when the servo information unnecessary process is performed on the servo information portion in which the abnormal position signal of FIG. 6 is generated. Since the sampling of the defective servo information section is stopped, the abnormal position signal does not occur at the defective servo information section, and no subsequent transient response occurs. Therefore, the rewrite operation ends successfully and is reported as a recoverable error.

[0020]

According to the present invention, the medium defect occurring in the servo information portion can be tolerated and the yield of HDA can be improved. Further, an unrecoverable error caused by a medium defect and TA of the MR head can be relieved as a recoverable error.

[Brief description of drawings]

FIG. 1 is a diagram showing servo information written on a disk in an embedded servo system.

FIG. 2 is an enlarged view of a servo information section.

FIG. 3 is a block diagram showing a configuration of a magnetic disk device.

FIG. 4 is a conceptual diagram of AM crushing processing.

FIG. 5 is an enlarged view of a servo information section after AM crushing processing.

FIG. 6 is a diagram showing an example of an abnormal position signal waveform due to a defect in a burst portion.

FIG. 7 is a diagram showing an example of a position signal waveform after performing a servo information unnecessary relief process.

FIG. 8 is a diagram showing a detailed sequence of servo information unnecessary relief processing.

[Explanation of symbols]

1 Spindle motor driver 2 VCM driver 3 R / W channel 4 R / W control circuit 5 Servo control circuit 6 Servo microcomputer 7 Drive control circuit 8 SCSI control circuit 9 Servo data detector 10 Peak hold circuit

─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Koichi Ono 2880, Kozu, Odawara-shi, Kanagawa Hitachi Storage Systems Division (72) Inventor Katsuhiro Tsuneda 2880, Kozu, Odawara, Kanagawa Hitachi Storage Systems Business Inside (56) References JP-A-8-190703 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) G11B 20 / 18,5 / 09

Claims (2)

(57) [Claims] 1. A servo control method in a magnetic disk device for generating a position signal based on servo information written on a medium surface to perform positioning control, wherein each servo information portion on a medium is sequentially read, A servo control method characterized in that it is determined whether or not there is an abnormality in the read servo information of the servo information section, and when it is determined that there is an abnormality, a part of information in the servo information section having the abnormality is erased. . 2. In a magnetic disk device that generates a position signal based on servo information written on a medium surface and performs positioning control, there is an abnormality in the servo information of each servo information section on the medium that is sequentially read. A magnetic disk device comprising: a means for determining whether or not there is an abnormality; and a means for erasing a part of information in the servo information portion having the abnormality when it is determined that there is an abnormality.