JP2001093104A - Operation error recovering method of magnetic disk device, and magnetic disk device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an operation error recovering method of a magnetic disk device capable of preventing the deterioration of the performance as a whole device and improving the error correction rate in the magnetic disk device utilizing a ramp load system, by improving the error retry system affected by the noise generated from the unbalance of a magnetic domain of an MR sensor. SOLUTION: By this operation error recovering method of the magnetic disk device and its device, the judgement is made whether the operation error is caused by the abnormality of servo data or not, and when this operation error is judged as caused by the abnormality of the servo data, a head is unloaded to the ramp to execute the process for correcting the unbalance of the magnetic domain of the MR sensor, and when the judgement is made that the operation error is not caused by the abnormality of the servo data, the head is moved to the landing area to execute the dummy write.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a magnetic disk drive using a ramp load method, and more particularly to a technique for restoring the function of an MR sensor of a magnetic disk drive using a magnetoresistive sensor (hereinafter referred to as an MR sensor) as a read head.
It relates to an error retry processing method.

[0002]

2. Description of the Related Art Conventionally, an MR sensor using an anisotropic magnetoresistance effect has been used as a reproducing head of a magnetic disk drive. In addition, a high magnetic resistance change is exhibited even with a low magnetic field stored in a medium, and a high recording density can be realized.

Generally, a GMR sensor has two ferromagnetic layers sandwiching a non-ferromagnetic layer. One ferromagnetic layer is called a free layer, and its magnetization direction is constant in a state without an external magnetic field, but can freely rotate in that direction when an external magnetic field is applied. The other ferromagnetic layer is referred to as a pinned layer, and its magnetization direction is fixed at 90 degrees with the magnetization direction of the free layer in the absence of an externally applied magnetic field, and does not change with an external magnetic field. In order to fix the magnetization direction of the pinned layer, an antiferromagnetic layer called a fixed layer is attached so as to be in direct contact with the pinned layer. The fixed layer provides a fixed magnetic field for pinning the magnetization direction of the pinned layer by exchange coupling. When the sensor is brought close to the surface of the disk, the magnetization direction of the free layer changes according to the direction of the magnetic flux on the disk surface and the electrical resistance of the sensor also changes, so if a bias current is supplied to the sensor in advance, the magnetization direction The data stored on the disk can be read as a change.

In order to magnetize the antiferromagnetic layer (fixed layer) in a predetermined direction, the antiferromagnetic layer is heated to block (Blocki).
ng) Cool above temperature and in the presence of a magnetic field suitable for pinning. The blocking temperature is a temperature at which the exchange anisotropy of the antiferromagnetic layer disappears. When the temperature of the GMR sensor used in the magnetic disk device is subjected to a long time temperature stress due to an increase in the temperature inside the disk device, physical contact between the GMR sensor and the magnetic disk serving as a storage medium, and heating by a bias current, etc. , The pinned layer cannot be pinned in the initial direction due to the influence of the magnetic field existing around it, and the amplification factor decreases or the reproduced waveform is deformed, so that the predetermined performance cannot be exhibited and the user data or servo data is read. An operation error occurs in connection with the error.

In a conventional magnetic disk drive which does not use the ramp load method, as shown in FIG. 2, a contact start / stop (CSS) area 22 or an inner / outer area where the head comes in contact with the head when the disk is stopped. Guard band (IGB) area 23, outer guard band (OGB) area 25 on the outer periphery
There is an area not used as the data area 24.
When a read error occurs in such a magnetic disk device, a seek operation of the head is executed, and an area without user data, that is, the CSS area 22 is read.
Thus, a write operation (hereinafter referred to as a dummy write operation) can be performed. In the CSS area 22, the bias current can be changed.

Here, the dummy write will be described.
A dummy write is similar to a normal user data write operation, in which servo data radially arranged on the disk surface at predetermined intervals is read via a magnetic head, and the head is moved to a predetermined track. This is a special write operation for performing positioning and executing a write operation only in a predetermined data area as shown in FIG. Therefore, the dummy
Write can be executed only when servo data can be read.

[0007] Even in the case where servo data cannot be read, in a conventional magnetic disk drive that does not use a ramp load method, a head is driven by driving a voice coil motor (hereinafter referred to as VCM). Was pressed against the CSS area 22 arranged on the innermost circumference, and a write current could be forcibly applied to the write head (hereinafter, referred to as forced write).

The forced write must be performed in an area on the disk where servo data and user data are not recorded. The reason is that in order to force a write, the user data area, IGB area,
This is because if the forced write is performed by mistake in the OGB area or the like, all the data 34 (including the servo data) on the disk will be erased.

Further, in the conventional magnetic disk drive using the ramp load method, as shown in FIG.
The S area 22 does not exist, only the IGB area 32 is arranged, and a landing area 34 exists on the outer periphery.

A problem in the case where forced writing is performed on a disk used in a magnetic disk device using such a ramp load method will be described below. In a disk used in a magnetic disk device using a ramp load method, a CSS executed when a state in which servo data cannot be read normally occurs in a disk used in a magnetic disk device not using a ramp load method. The same processing as the forced writing in the area 22 cannot be performed. As shown in FIG. 3, although there is a margin space on the inner periphery of the IGB area 32, the physical space is smaller than that of the CSS area 22 used in the conventional method, and there is a manufacturing variation. Even if the head is pressed to the innermost circumference by driving the VCM due to the
It may involve the GB area 32, and the execution of the forced write on the innermost circumference is dangerous.

A problem in executing a dummy write on a disk used in a magnetic disk device using the ramp load method will be described. In the disk used in the magnetic disk device using the ramp load method, as shown in FIG. 3, since there is no CSS area 22, dummy writing on the inner periphery can be performed only in the IGB area 32. Since the radial position is shifted more inward than the disk used in the magnetic disk drive that does not use the ramp load method, there is a high possibility that the disk is deformed due to the disk holder. Therefore, there is a high possibility that the dummy write fails.

[0012]

The performance of a GMR sensor mounted on an actuator arm of a magnetic disk drive can be known through an operation error display of a running device. However, an operation error also occurs due to a cause other than a decrease in the performance of the GMR sensor, such as a shake of the spindle or a vibration of the apparatus. Errors based on the degradation of the performance of the GMR sensor are recognized inside the apparatus together with errors based on other causes and processed by an error recovery program (ERP). G
When the performance of the MR sensor deteriorates, the ERP attempts to recover the error by reassigning a sector that has become unreadable and cannot be written to a replacement sector. However, the number of replacement sectors is limited, and when the replacement sector is used up, The error cannot be recovered by reassignment, and the specification of the substitute sector is not preferable because it lowers the head access speed of the magnetic disk.

[0013] Forcibly write, dummy, and write in a disk used in a magnetic disk device using a ramp load system.
If the conventional method is used as it is, the data may be destroyed, and the conventional method is not preferable.

Here, an object of the present invention is to improve an error retry method caused by noise generated due to imbalance of magnetic domains of an MR sensor serving as a read head in a magnetic disk drive using a ramp load method. ,
It is an object of the present invention to provide a method for recovering an operation error of a magnetic disk device in order to reduce the performance of the magnetic disk device as a whole and improve an error correction rate, and to provide a magnetic disk device to which this method is applied.

[0015]

In order to achieve the above object, an operation error recovery method for a magnetic disk drive according to the present invention comprises a storage medium having a data area, a landing area and a servo area disposed on a surface thereof, an MR sensor, A head using the same, an actuator arm for mounting the head and positioning the head at a predetermined position on the surface of the storage medium based on servo information in the servo area, and controlling an operation and a read / write operation of the actuator arm And an operation error occurring in a magnetic disk device using a load / unload method in which the head is engaged with a ramp disposed at a position distant from the surface of the storage medium when the storage medium is stopped. A method for recovering, wherein the operation error is caused by an abnormality of the servo data. Determining whether the operation error is due to an abnormality in the servo data, and unloading the head to the ramp in order to recover the operation error. Performing a process for correcting imbalance of the magnetic domains of the MR sensor; and, when the determining step determines that the operation error is not caused by an abnormality in the servo data, recovering the operation error. Moving the head to the landing area and performing a dummy write.

According to the method of recovering an operation error occurring in the magnetic disk drive configured as described above,
Since the error retry method caused by noise generated due to the magnetic domain imbalance of the MR sensor serving as the read head is improved, the performance of the magnetic disk device as a whole deteriorates.
The error correction rate can be improved.

Further, according to the magnetic disk drive of the present invention, there is provided a storage medium having a data area, a landing area, and a servo area disposed on a surface thereof, a head using an MR sensor, the head mounted thereon, and the head mounted on the servo. An actuator arm positioned at a predetermined position on the surface of the storage medium based on the servo information of the area, and a control unit that controls the operation and read / write operation of the actuator arm, and controls the head when the storage medium is stopped. A magnetic disk drive using a load / unload method that engages with a ramp disposed at a position distant from the surface of the storage medium, wherein the control unit or the storage medium detects that the operation error is caused by an abnormality in the servo data. Judge whether the operation error is caused by an abnormality of the servo data. If it is determined that there, in order to restore the operation error, to unload the head to the lamp, to execute a process to correct the imbalance of magnetic domains of the MR sensor. Further, when it is determined that the operation error is not due to an abnormality in the servo data, the head is moved to the landing area and a dummy write is executed in order to recover the operation error. A magnetic disk drive that stores a program readable by a host system that causes the host system to execute an operating method.

Also in the magnetic disk device thus configured, the error retry method caused by the noise generated due to the imbalance of the magnetic domains of the MR sensor serving as the read head is improved, so that the performance of the magnetic disk device as a whole is improved. The reduction and the improvement of the error correction rate can be achieved.

[0019]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A magnetic disk drive according to an embodiment of the present invention will be described with reference to FIGS. FIG. 3 is a view for explaining a medium used in a magnetic disk drive using a ramp load method employed in an embodiment of the present invention. As shown in FIG. 3, the medium held in the magnetic disk device using the ramp load method does not have the CSS area 22 of the medium used in the magnetic disk device not using the ramp load method. The IGB area 32 is arranged on the inner periphery of the medium 43, and the data area 33 is arranged on the outer side in the radial direction. Further, the head 42a moves from the ramp 45 on the outer periphery thereof, and the landing area 34 is arranged as the smallest area. Servo data, which is positioning information of the head 42 a, is provided on the medium 43 at a predetermined distance radially in the radial direction of the medium 43.
It has an area 31. Servo data is in IGB area 32
To the middle of the landing area 34.

Here, when the head 42a descends from the ramp 45 and moves onto the medium 43, it is called a load.
The operation in which the VCM 44 rides on the ramp 45 from the inner peripheral side to the outer peripheral side of the medium 43 as in the direction indicated by the arrow B and the position indicated by the dotted line in FIG. 4 is called unloading.

FIG. 4 is a schematic diagram for explaining a head (or hard) disk assembly 41 (hereinafter, referred to as HDA 41) mainly composed of mechanical mechanism components of the magnetic disk drive, and FIG. 5 is a magnetic disk. Printed circuit board 5 on which electronic components for controlling the HDA 41 of the apparatus are mounted
1 (hereinafter, referred to as PCB 51). The HDA 41 and the PCB 51 are electrically connected by an interface 48, and exchange various signals with various circuits in the PCB 51 to control each mechanism in the HDA 41.

First, the HDA 41 of the magnetic disk drive to which the embodiment of the present invention is applied will be described with reference to FIG. A plurality of media 43 are stacked on a spindle shaft (not shown), and when writing or reading data, a spindle motor 46 is used.
(Hereinafter, referred to as SPM 46) in a fixed direction together with the spindle shaft. Each medium 43 includes a data area 33 used for recording information and a medium 43.
And a landing area 34 which is a non-recording area located at a position on the disk 43 immediately before the head 42a is retracted to the ramp 45 when the rotation of the head 42 is stopped. The data area 33 is divided into a plurality of annular tracks. Each track has a synchronization signal, user data composed of 512 bytes, and an Ecc (Error Correction Code).
e) are divided into a plurality of data sectors.
Further, in the data area 33, a servo area 31 in which servo data is recorded as a plurality of blocks in a discrete area is arranged. Multiple media 43
Uses the front and back sides as recording areas, respectively.
One cylinder is formed by a plurality of tracks on a plurality of recording surfaces included in a cylindrical shape including one track.

The actuator arm 42b is driven by a voice coil motor 44 (hereinafter referred to as VCM 44), and rotates the surface of the medium 43 in the directions of arrows A and B. The actuator arm 42b has a laminated structure corresponding to a plurality of media 43, and a plurality of heads 42a are provided at the respective leading ends thereof through flexible members having elasticity applied to the upper and lower surfaces of the respective media 43. Each is installed. The magnetic head 42a includes a slider, a read-only GMR sensor (not shown) mounted on the slider, and a write-only inductive transducer (not shown). The GMR sensor reads servo data when reading, writing and seeking data, and reads user data when reading. AE (Arm elec
tronics) module (not shown)
A bias current is supplied to the arm 42b, the bias current is supplied to the GMR sensor, the data stored in the magnetic disk is read, the read signal generated by the head 17 is amplified, and a write current is supplied to a write converter. Is controlled.

When data is written or read, the medium 43 on which the actuator arm 42b is rotating is rotated.
Rotating on the surface of the GMR sensor and the data area 33
Perform a seek operation for operating an arbitrary track position on the surface of the. At this time, the slider is given a floating force by the airflow generated on the surface of the medium 43, and the head 43a floats at a certain distance from the surface of the medium 43, and that distance is maintained.

When the medium 43 is stopped, the head 42a is moved away from the medium 43 and stopped.
A lamp 45 is disposed at a position along the outer periphery of the medium 43 in the first medium 1.

Next, the structure and operation of the PCB 51 of the magnetic disk drive to which the embodiment of the present invention is applied will be described with reference to FIG. FIG. 5 shows a printed circuit board circuit 51 (hereinafter, referred to as a printed circuit board) on which electronic components for controlling the HDA 41 of the magnetic disk drive are mounted.
Called PCB51. FIG. CPU
52 controls the entire magnetic disk device according to a program stored in the ROM 58. The gate array 53 includes a read / write circuit 54 (hereinafter, referred to as an R / W channel 54), a logic circuit for controlling the SPM / VCM control circuit 55, and servo data transmitted from the R / W channel 54. Check that the information is correct. If there is an error in the servo data, the CPU 52 can be interrupted. The hard disk controller 56 (hereinafter, referred to as HDC 56) interfaces with a host computer 59 at a higher level. The RAM 57 is used as a buffer for user data and as a memory for the CPU 52. The ROM 58 stores a program for controlling the CPU 52. All retry control procedures are also recorded in the ROM 58. R / W channel 54
During reading, analog data read from the head amplifier 47 is digitized and sent to the HDC 56. At the time of writing, the data sent from the HDC 56 is modulated and sent to the head amplifier 47. Head amp 4
AGC circuit (not shown) that amplifies the output from 7 to a constant amplitude, vertical and non-symmetrical correction circuit (not shown) that automatically corrects vertical and non-symmetrical of the output waveform, partial response (Part
and various filters for performing equalization for performing ial response). Spindle motor / voice coil motor control circuit 55 (hereinafter SPM / VCM
Called control circuit 55. ) Controls the rotation of the SPM 46 and the VCM 44 to control the movement of the head 42a and the load / unload of the head 42a.

The HDA 41 and the PCB 51 are electrically connected by an interface 48, and exchange various signals with various circuits in the PCB 51 to control each mechanism in the HDA 41.

The CPU 52 includes an actuator arm 42
b) and controls the execution of data read / write operations. The seek operation of the actuator arm 42b is controlled through the SPM / VCM control circuit 51,
Further, the read / write operation of the head 42a is controlled through the R / W channel 54. R / W channel 54 is HD
Connected to C56. The R / W channel 54 is a read / write channel for bidirectionally converting a digital signal of user data and a current signal supplied to the head 43 and generated by the head 43. In this embodiment, a partial response (Partial Response) channel is employed.

The HDC 39 is connected to a memory (RAM) 57 and a host computer 59 via a host interface 48. The HDC 56 controls the RAM 57, transfers data between the RAM 57 and the medium 43,
It creates an ID table and generates a position error signal (PES: Position Error Signal) from the servo data, and sends the position information of the head 43 to the CPU 52. RAM
57 stores a microcode for control of the device, an ID table, and the like read from the medium 43 when the device is started. HDC 56 is connected to CPU 52 and CP
U52 has a memory (ROM) 58 and a gate array 53
Are further connected, and an SPM / VCM control circuit 55 is connected via the gate array 53. C
The PU 52 interprets the command sent from the host computer 59 and causes the HDC 56 to create an ID table, further instructs the HDC 56 to read and write data from and to the address specified by the command, and also causes the head 42a to operate at a predetermined level. The control information for seeking to the track
Based on the head position information generated in 56, the SPM /
It is sent to the VCM control circuit 55.

The SPM / VMC control circuit 55
The VCM 44 is driven based on the control information from the CPU 52 so as to position the head 42a on a predetermined track. The ROM 58 stores micro codes necessary for starting the apparatus. A program for executing the present embodiment is read from the medium 43 to the RAM 57 as a part of the ERP when the apparatus is started, and when an error occurs in the apparatus, the CPU 52 writes the ERP according to the operation mode of the apparatus at that time. , And seek ERP in a predetermined procedure.

When user data is written to each sector of the medium 43, an HDC 56 calculates an ECC used for detecting and correcting a data error generated according to a bit pattern of the data, and writes the ECC to the medium 43 together with the user data. Written automatically. At the time of reading, even if the correct sector address is accessed, an error occurs in the bit of the read user data, and the HDC 56
If the data cannot be reproduced correctly using the ECC, a reading error occurs. At the time of writing, it is necessary to perform servo control for positioning the write converter of the head at the write center position of each track and in the designated sector while reading servo data with the GMR sensor. However, since the servo control does not work properly, the head 42a cannot be positioned in the cylinder within a predetermined time to be in a writable state, and furthermore, if the writing cannot be performed even if the address of the writing sector is reassigned to the alternative sector, the writing is performed. I get an error. Reassignment is based on Automatic Defect Re-allo (ADR) in ERP.
cation) procedure, and if the reassignment is successful, it will not cause an error in the device. If it is not possible to position the head 42a on a predetermined cylinder within a predetermined time because correct servo data cannot be read during a seek, a seek error occurs.

More simply, in FIG. 4, the signal read from the medium 43 by the head 42a is amplified by a head amplifier 47, and then amplified by a P shown in FIG.
It is sent to the R / W channel 54 on the CB 51. R
The / W channel 54 digitizes the signal read from the head 42a and sends it to the HDC 56. HDC65 is R
It is determined whether there is no error in the signal sent from / W channel 54. If an error is detected in HDC 56, C
The PU 52 re-reads the sector in which the error has occurred in accordance with a procedure previously described in the ROM 58, and attempts to recover from the error. Due to the magnetic domain imbalance of the head 42a, the head 4a
If an error has occurred due to the noise generated in 2a, the error cannot be recovered by ordinary error recovery means. Here, the normal error recovery means indicates a change in the setting of the R / W channel 54, a response to off-track write, and a read / retry. Therefore, if the error cannot be recovered even by performing the normal error recovery means, it is determined that the magnetic domain imbalance of the read head has occurred, and the operation for recovering the magnetic domain of the read head is performed. First, a process for not erasing user data and servo data, and a process for terminating the process in as short a time as possible so as not to cause a decrease in performance due to retry, are executed with priority.

Next, an operation error recovery method according to an embodiment of the present invention will be described with reference to flowcharts shown in FIGS. First, when a read error occurs, the initial value of the retry counter n is set to 0 (step S).
1) A normal error recovery operation (retry operation) including a change in the setting of the R / W channel 54 and a response to off-track write is executed (step S2). next,
During normal error recovery operation (retry operation), servo
It is determined whether an error has occurred in the data (whether the servo data has been read normally) (step S).
3). If the servo data cannot be read normally, the gate array 53 decodes the servo data sent from the R / W channel 54 and sends the data to the CPU 52. The CPU 52 executes an interrupt process when the data is equal to or greater than the servo data shown in FIG. 7 based on the information of the received servo data (step S5). After the end of the interrupt processing, the continuation of the normal retry processing is executed.

In the interrupt processing (step S5), first, it is determined that there is an error in the servo data (step S21). If it is determined that there is an error in the servo data, a dummy write on the medium 43 is performed. , There is a possibility that data will be erased in the forced write process,
It is necessary to unload the head 42a to the ramp 45 and perform the forced write process, even if the processing time of the entire magnetic disk device is reduced. First, the head 42a is unloaded to the lamp 45 (step S22), and any one of a forced write process, a change in bias current to the GMR element, and a reset of read channel parameters, or a plurality of processes is executed. Then, an operation of eliminating the magnetic domain imbalance of the head 42a is executed (Step S23). Also, at this time, since the read signal is not output at all during the adjustment of the GMR bias current, the head 42a
Can no longer be positioned, so that the head 4
2a may collide. Therefore, it is necessary to unload the head 42a to the ramp 54 for execution of the bias current fluctuation even at the expense of shortening the processing time of the entire magnetic disk device. Then, the head 42
a is loaded on the medium 43 again (step S2).
4), end the interrupt processing.

If it is determined in step S21 that there is no error in the servo data, this interrupt processing is interrupted, and the processing shifts to step S6 of the main routine (FIG. 6). In step S3 of FIG. 6, if no error is detected in the normal read retry operation, it is determined that the error correction has been successful, and the retry process ends (step S4).

In FIG. 6, after the interrupt processing is completed, the head 42a is positioned on the medium 43, so that a dummy write operation is executed in the landing area 34 (step S6). Thereafter, a normal error retry operation is executed again (step S7), and it is verified whether the error has been recovered (step S8). Step S8
When it is determined in the verification of that the error has been recovered, the error collection ends and the collection is successful (step S10), and a series of processing ends. If it is determined in the verification in step 8 that the error has not been recovered, it is determined whether the counter n of the number of retries has exceeded a predetermined maximum value (MAX retry value) (step S9). When it is determined that the counter n of the number of retries has exceeded a predetermined maximum value (MAX retry value), it is determined that an uncorrectable error (uncorrectable error) has occurred, and the collection process ends (step S11). If it is determined in step S9 that the counter n of the number of retries has not exceeded the predetermined maximum value (MAX retry value), the processing is performed in step S2.
Then, the process proceeds to the normal error / retry operation, and the retry is executed.

In the above-described interrupt processing, an interrupt is always applied during the operation of the magnetic disk drive, and the information of the servo data cannot be read normally regardless of the rotation state of the disk. Is executed.

[0038]

As described in detail above, according to the operation error recovery method and the magnetic disk drive of the present invention, the read / write
The error retry method caused by noise generated due to the imbalance of the magnetic domains of the MR sensor as the head is improved, and the performance of the entire magnetic disk device is reduced and the error correction rate can be improved.

[Brief description of the drawings]

FIG. 1 illustrates a write operation for explaining a dummy write operation.
FIG. 3 is a conceptual diagram of gate signal generation timing and a track format.

FIG. 2 is a conceptual diagram of a disk format in a contact start / stop (CSS) type HDD.

FIG. 3 is a conceptual diagram of a disk format in a HDD of a type in which a head is loaded / unloaded (ramp loaded) on the outer peripheral side of the disk.

FIG. 4 is a conceptual diagram of a hard disk assembly (HDA) in the HDD of the embodiment of the present application.

FIG. 5 is a conceptual diagram illustrating a circuit configuration on a PCB in the HDD according to the embodiment of the present application.

FIG. 6 is a flowchart illustrating a retry operation according to the embodiment of the present application.

FIG. 7 is a flowchart illustrating an interrupt process when a servo error occurs according to the embodiment of the present application.

[Explanation of symbols]

 41 Hard disk assembly (HDA) 42a Head 42b Actuator arm 43 Medium 44 Voice coil motor (VCM) 45 Lamp 46 Spindle motor (SPM) 47 Head amplifier 51 PCB 52 CPU 53 Gate array 54 Read / write channel 55 SPM / VCM control circuit 56 Controller (HDC) 57 RAM 58 ROM 59 … Host system

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G11B 20/18 576 G11B 20/18 576Z

Claims (3)

[Claims] 1. A storage medium having a data area, a landing area, and a servo area disposed on a surface thereof; a head using an MR sensor; and a head mounted with the head, and the head mounted on the storage medium based on servo information in the servo area. An actuator arm positioned at a predetermined position on the surface of the storage medium, and a control unit for controlling the operation and the read / write operation of the actuator arm, wherein the head is located at a position away from the storage medium surface when the storage medium is stopped. The load that engages the placed lamp
A method for recovering an operation error occurring in a magnetic disk device using an unloading method, comprising: a step of determining whether the operation error is due to an abnormality in the servo data; and If it is determined that this is due to an abnormality in the servo data, in order to recover the operation error, a process of unloading the head to the ramp and correcting the magnetic domain imbalance of the MR sensor is performed. Performing the operation, if it is determined by the determination step that the operation error is not due to an abnormality in the servo data, to recover the operation error, move the head to the landing area, the dummy write Performing the steps. 2. The process for correcting the magnetic domain imbalance of the MR sensor comprises the step of executing at least one of a forced write operation, resetting of an MR bias current, and resetting of a read / write channel parameter. The method of claim 1. 3. A storage medium having a data area, a landing area, and a servo area disposed on a surface thereof, a head using an MR sensor, the head mounted, and the head storing the head based on servo information of the servo area. An actuator arm positioned at a predetermined position on the surface of the medium, and a control unit for controlling operation and read / write operation of the actuator arm, wherein the head is separated from the surface of the storage medium when the storage medium is stopped. Load that engages the ramp
A magnetic disk drive using an unloading method, wherein the control unit or a storage medium stores a program readable by a host system that causes the host system to execute the method according to any one of claims 1 and 2. The magnetic disk device that stores.