Magnetic disk device

Abstract

A disk device is provided to improve the performance of multiple actuators. The disk device according to the embodiments has two or more independently driveable actuator blocks, performing low-jerk seek control with limited jerk (i.e., accelerometer). The disk device is characterized in that, when the first actuator block does not access a data sector of the disk, a second actuator block, not the first actuator block, accesses a data sector of the disk through high-jerk seek control.

Description

[0001] This application enjoys priority based on Japanese Patent Application No. 2022-012108 (filed on January 28, 2022). This application incorporates the entire contents of that basic application by reference. Technical Field

[0002] Embodiments of the present invention relate to disk drives. Background Technology

[0003] Multi-actuator disk drives have multiple actuators that can be driven independently. When any actuator performs a seek operation, it excites the vibration of other actuators, causing the tracking head to shake. Sometimes, the position error (PES) of the tracking head relative to the target position can increase. Therefore, control is implemented to limit the jerk (acceleration) during actuator seek.

[0004] However, when Jerk is restricted, seek time becomes slower, and sometimes command access performance (the number of commands that can be accessed per unit of time) decreases. Summary of the Invention

[0005] This embodiment provides a disk device for improving the performance of multiple actuators.

[0006] The disk device involved in the implementation is a disk device having two or more independently driveable actuator blocks and performing low-jerk seek control with limited jerk (acceleration), characterized in that, when the first actuator block does not access the data sectors of the disk, the second actuator block, which is not the first actuator block, accesses the data sectors of the disk through seek control with high jerk. Attached Figure Description

[0007] Figure 1 This is a configuration diagram of the disk device involved in the implementation method.

[0008] Figure 2 (a) and (b) are schematic diagrams illustrating the configuration of the actuator block and disk, etc., according to this embodiment.

[0009] Figure 3 This is a block diagram illustrating the functional configuration of the servo control unit of the disk device according to this embodiment.

[0010] Figure 4 This diagram illustrates an example of the effect of the actuator of the disk device according to this embodiment on other actuators when the actuator is started.

[0011] Figure 5 This is a diagram illustrating the data flow in the disk device according to this embodiment, under the function of suppressing the influence of vibration interference between actuators.

[0012] Figure 6 This is a diagram illustrating the reordering table involved in this implementation.

[0013] Figure 7 This is a timing diagram of the disk device performing seek operations according to this embodiment.

[0014] Figure 8 This is a flowchart illustrating the command selection process for the disk device involved in this implementation.

[0015] Figure 9 This is a flowchart illustrating the type of Jerk seek control set for the disk device involved in this embodiment.

[0016] Figure 10 This is a flowchart illustrating the type of disk device setting rearrangement in this implementation.

[0017] Figure 11 This is a flowchart illustrating the command selection process for the disk device involved in this implementation.

[0018] Figure 12 This is a flowchart illustrating the seek control process performed by the disk device involved in this embodiment.

[0019] Figure 13A (a) to (d) are diagrams showing the timing of seek control performed by the two control systems of the disk device involved in this embodiment.

[0020] Figure 13B This is a diagram showing the relationship between the operating states of the two control systems of the disk device involved in this embodiment.

[0021] Figure 14 This is a diagram illustrating an example of the relationship between acceleration and position in seek control of a disk device according to this embodiment.

[0022] Figure 15 This is a schematic diagram of the disk device involved in the variation example.

[0023] Label Explanation

[0024] 1. Disk drive; 2. Host system (host); 10. HDA; 12. Spindle; 13. Spindle motor (SPM); 20A. Driver IC; 30A. Header amplifier IC; 40A. Read / write (R / W) channel; 50A. Microprocessor (MPU); 60A. Hard disk controller (HDC); 70A. Volatile memory; 80A. Buffer memory; 90A. Non-volatile memory; 100A. Actuator block; 170. Shock sensor; 180. Write inhibit detector; 190. Controller communication unit; 210A. SPM control unit; 220A. VCM control unit; 230A. MA Control Unit; 310A Read Head Selection Unit; 320A Read Signal Detection Unit; 410A Write Inhibition Unit; 411A Impact Sensor Write Inhibition Decision Unit; 412A HDC Write Inhibition Decision Unit; 610A Command Control Unit; 611A Actuator Status Confirmation Unit; 612A Actuator Status Communication Unit; 613A Command Selection Unit; 620A Servo Control Unit; 621A Tracking Control Unit; 622A Seek Control Unit; 630A Write Action Decision Unit; 631A Position-Based Write Action Decision Unit; 632A Speed-Based Write Action Decision Unit; 6131A Reordering List Selection Unit; 6132A Predicted Seek Time Limit Command Selection Unit; 6221A Low Jerk Seek Control Unit; 6222A High Jerk Seek Control Unit; 6223A JIT seek control unit; DK1 disk; AC1 actuator; AM11 arm; HD11 head; MA11 miniature actuator; RH11 read head; WH11 write head. Detailed Implementation

[0025] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0026] (Implementation Method)

[0027] This embodiment illustrates an example of seek control for improving the command access performance of a disk device with two actuators. Command access performance can also be the number of commands that can be processed per unit time.

[0028] For example, when one actuator (let's call it actuator A) performs a seek operation, it excites the vibration of the other actuator (let's call it actuator B), causing the head of actuator B, which is tracking, to wobble. As a result, the position error signal (PES) of actuator B's tracking head relative to the target position increases, potentially leading to errors in reading and writing data sectors. Therefore, it is proposed that by limiting the jerk during actuator A's seek operation, the wobble of actuator B's tracking head can be reduced during data sector reads and writes. However, limiting the jerk slows down the seek time, degrading the command access performance of actuator A. This embodiment illustrates an example of seek control that improves command access performance under such circumstances.

[0029] Figure 1 This is a configuration diagram of the disk device involved in the implementation method.

[0030] The disk device 1 is, for example, a hard disk drive (HDD), which is a multi-actuator disk device having two actuator blocks 100A and actuator block 100B.

[0031] The disk drive 1 in this embodiment includes two control systems, A and B. Control system A controls actuator block 100A to control access to disk DK1. Control system B controls actuator block 100B to control access to disk DK2. Control systems A and B are capable of data communication. Regarding the functions of control systems A and B, function blocks with the same name are distinguished by adding A and B to the reference numerals in the accompanying drawings.

[0032] HDA10 is a headstock assembly, housing multiple discs, multiple actuator blocks, a spindle, etc. HDA10 includes at least two actuator blocks 100 and discs DK.

[0033] Disk DK1 and disk DK2 (collectively referred to as disk DK unless otherwise distinguished) are disks that store data and are controlled by control systems A and B respectively.

[0034] The main shaft 12 is the support for disks DK1 and DK2, and is installed in the housing HS, etc.

[0035] The spindle motor (SPM) 13 is equipped with a spindle 12, which rotates the spindle 12.

[0036] Actuator block 100A and actuator block 100B (collectively referred to as actuator block 100 unless otherwise distinguished) are controlled by control systems A and B respectively, and perform data reading and writing on different disks DK1 and DK2.

[0037] Figure 2This is a schematic diagram illustrating the configuration of the actuator block and disk, etc., involved in this embodiment.

[0038] Figure 2 (a) is a schematic diagram of actuator block 100 (actuator block 100A or actuator block 100B). When the names of the constituent elements in actuator block 100A and actuator block 100B are the same, the constituent elements are identical. Therefore, the reference numeral number of actuator block 100B is shown in parentheses following the reference numeral number of actuator block 100A.

[0039] Actuator AC1 is controlled by control system A and is a voice coil motor (VCM) type actuator, mounted on coaxial BR.

[0040] Actuator AC2 functions the same as actuator AC1 in control system B. Unless otherwise specified, actuators AC1 and AC2 are collectively referred to as actuator AC.

[0041] Arms AM11 and AM12 (collectively referred to as arm AM1 unless otherwise specified) are connected to actuator AC1 and head HD1 via a clamping disc DK1, and support head HD1. Arm AM1 is controlled by control system A.

[0042] Arms AM21 and AM22 (collectively referred to as arm AM2 unless otherwise specified) are connected to actuator AC2 and head HD2 via clamping disc DK2, supporting head HD2. Arm AM2 is controlled by control system B in the same manner as arm AM1. Arms AM1 and AM2 are collectively referred to as arm AM unless otherwise specified.

[0043] Multi-actuator MA11 and multi-actuator MA12 (collectively referred to as multi-actuator MA1 unless otherwise distinguished) are actuators that are respectively mounted on the suspensions (not shown) of arms AM11 and AM12, clamping disc DK1, and controlled by control system A to control head HD11 and head HD12, respectively. More specifically, the micro-actuator MA1 controls the radial movement of head HD1 of disc DK1 with finer control than the radial movement control of head HD1 via voice coil motor VCM. The micro-actuator MA1 can also be driven independently of VCM.

[0044] Multi-actuator MA21 and multi-actuator MA22 (collectively referred to as multi-actuator MA2 unless otherwise distinguished) are controlled by control system B in the same way as multi-actuator MA1. Multi-actuator MA1 and multi-actuator MA2 are collectively referred to as multi-actuator MA unless otherwise distinguished.

[0045] Heads HD11 and HD12 (collectively referred to as head HD1 unless otherwise distinguished) are respectively mounted on the front ends of arms AM11 and AM12, clamping disk DK1. Heads HD11 and HD12 are controlled by control system A to read and write data on the upper and lower surfaces of disk DK1, respectively.

[0046] Heads HD21 and HD22 (collectively referred to as head HD2 unless otherwise specified) are respectively mounted on the front ends of arms AM21 and AM22, clamping disc DK2. Head HD2 is controlled by control system B in the same way as head HD1. Heads HD1 and HD2 are collectively referred to as head HD unless otherwise specified.

[0047] Each head HD is mounted on a slider (not shown), which is mounted on the suspension of the arm AM. The head HD is selected and operated by the head selection unit (also called the read head selection unit 310) described later.

[0048] The time taken for the DK (Device Kinker) to rotate by an angle α (when the rotation direction of the DK is set to clockwise) to move the target data sector TGT to the position of the head HD is called the rotational time. Specifically, the rotational time from when the head HD seeks the track containing the target data sector TGT (called the target track) is called the rotational latency. More specifically, when the SPM angle (or servo sector number, etc.) of the current position of the head HD on the disk is set to α1, and the SPM angle (or servo sector number, etc.) of the target sector TGT is set to α2, the time required to rotate the difference α between these two SPM angles (or servo sector number, etc.) is equivalent to the rotational time or rotational latency. Furthermore, the time from the start of the head HD's seek to reaching the target data sector TGT is called the command access time.

[0049] When the head disk (HD) reaches the target track via seek, and the target data sector (TGT) passes the position of the head disk (DK) in the rotation direction of the disk (DK), the head disk waits for the target data sector (TGT) while the DK rotates. In this case, the command access time is calculated by adding the seek time from the current head position to the target data sector (TGT) to the rotation time or rotation wait time corresponding to the number of additional rotations. In the basic command rearrangement, the command with the shortest access time is selected.

[0050] The head HD has a write head WH for writing data to the disk DK and a read head RH for reading data written to the disk DK. Hereinafter, the process of writing data to the disk DK will sometimes be referred to as write processing, and the process of reading data from the disk DK will sometimes be referred to as read processing. Additionally, recording data to a predetermined recording area, reading data from a predetermined recording area, configuring the head HD at a predetermined position on the disk DK, writing data to a predetermined area of ​​the disk DK, and reading data from a predetermined area of ​​the disk DK will sometimes be referred to as access.

[0051] Read heads RH11 and RH12 (collectively referred to as read head RH1 unless otherwise specified) are located on head HD1. Read heads RH11 and RH12 are controlled by control system A to read data from the upper and lower surfaces of disk DK1, respectively.

[0052] Read heads RH21 and RH22 (collectively referred to as read head RH2 unless otherwise specified) are located on head HD2 and are controlled by control system B in the same manner as read head RH1. Read heads RH1 and RH2 are collectively referred to as read head RH unless otherwise specified.

[0053] Write heads WH11 and WH12 (collectively referred to as write head WH1 unless otherwise specified) are located on head HD1. Write heads WH11 and WH12 are controlled by control system A to write data to the upper and lower surfaces of disk DK1.

[0054] Write heads WH21 and WH22 (collectively referred to as write head WH2 unless otherwise distinguished) are located on head HD2 and are controlled by control system B in the same way as write head WH1. Write heads WH1 and WH2 are collectively referred to as write head WH unless otherwise distinguished.

[0055] Figure 2 (b) is a schematic diagram of a disk DK. A disk DK is a general-purpose disk, and detailed descriptions are omitted.

[0056] A disk drive (DK), for example, is a disk-shaped hard disk that uses magnetism to write data. The DK is mounted on the spindle 12 and rotated by the drive of the SPM 13. The direction along the circumference of the DK (its upper and lower surfaces) is called the circumferential direction, and the direction orthogonal to the circumferential direction of the DK (its upper and lower surfaces) is called the radial direction.

[0057] The disk DK is divided into multiple regions called tracks in the radial direction (or concentric circles) centered on the main axis 12. Additionally, the disk DK is divided into multiple regions called sectors in the circumferential direction. Figure 2(b) shows an example with one track TR and one sector SCT, with the target data sector TGT shown in the sector SCT on track TR.

[0058] In seek control, after the actuator block 100's arm AM moves to move the head HD onto the track TR, the head HD moves to the target data sector TGT by rotating the disk DK. Then, tracking control is implemented by a micro-actuator MA, etc. Tracking control is a general technique, and detailed information is not provided.

[0059] return Figure 1 The driver IC20A is a function of the control system A, which controls the various functions of HDA10 according to the control from MPU50A or HDC60A, etc.

[0060] Driver IC20B functions the same as driver IC20A in control system B. Unless otherwise specified, driver IC20A and driver IC20B are collectively referred to as driver IC20.

[0061] In addition, there is a case where the driver IC20 does not have a micro actuator MA, in which case the MA control unit 230 may also be absent.

[0062] The SPM control unit 210A is a function shared by control systems A and B, controlling the rotation of the spindle motor SPM13 of HDA10. The SPM control unit 210A... Figure 1 It can be set in driver IC20A, but it can also be set in driver IC20B.

[0063] The VCM control unit 220A controls the drive of the voice coil motor (VCM) by controlling the current (or voltage) supplied to it, which in turn controls the actuator block 100A of the HDA10.

[0064] VCM control unit 220B performs the same function as VCM control unit 220A in control system B. Without specifically distinguishing between VCM control unit 220A and VCM control unit 220B, they are collectively referred to as VCM control unit 220.

[0065] The MA control unit 230A controls the drive of the micro-actuator MA1 by controlling the current (or voltage) supplied to the micro-actuator MA1.

[0066] MA control unit 230B performs the same function as MA control unit 230A in control system B. Unless otherwise specified, MA control unit 230A and MA control unit 230B are collectively referred to as MA control unit 230.

[0067] The head amplifier IC30A, for example, is a preamplifier that amplifies the read signal read from disk DK1 and outputs it to the read / write (R / W) channel 40A. The head amplifier IC30A is electrically connected to head HD1 (head HD11, head HD12). Additionally, the head amplifier IC30A outputs a write current to head HD1 corresponding to the signal output from the R / W channel 40A.

[0068] Head amplifier IC30B functions the same as head amplifier IC30A in control system B. Unless otherwise specified, head amplifier IC30A and head amplifier IC30B are collectively referred to as head amplifier IC30. Head amplifier IC30 includes a read head selection unit 310 and a read signal detection unit 320.

[0069] The read head selection unit 310A selects the read head RH1 for reading data from disk DK1.

[0070] The read head selection unit 310B functions the same as the read head selection unit 310A in the control system B. Unless otherwise specified, the read head selection units 310A and 310B are collectively referred to as the read head selection unit 310.

[0071] The read signal detection unit 320A detects the signal (read signal) read from disk DK1 through the read head RH1.

[0072] The read signal detection unit 320B performs the same function as the read signal detection unit 320A in the control system B. Unless otherwise specified, the read signal detection units 320A and 320B are collectively referred to as the read signal detection unit 320.

[0073] R / W channel 40A performs signal processing for read data transferred from disk DK1 to host system 2 and write data transferred from host system 2, according to instructions from MPU50A. R / W channel 40A is electrically connected to head amplifier IC30A, MPU50A, HDC60A, and write inhibit detector 180, etc.

[0074] R / W channel 40B functions the same as R / W channel 40A in control system B. Unless otherwise specified, R / W channel 40A and R / W channel 40B are collectively referred to as R / W channel 40.

[0075] The write inhibit unit 410A specifies the inhibit (or stop) of writing (or writing operation) to disk DK1 via head HD1, and outputs control signals to head amplifier IC30A, etc.

[0076] The write inhibit unit 410B functions the same as the write inhibit unit 410A in the control system B. Without specifically distinguishing between the write inhibit unit 410A and the write inhibit unit 410B, they are collectively referred to as the write inhibit unit 410.

[0077] When the impact sensor write prohibition determination unit 411A receives a write prohibition determination execution signal generated by the write prohibition detector 180 based on vibration or impact detected by the impact sensor 170, it determines whether to prohibit (or stop) the writing (or writing operation) performed through the head HD1.

[0078] For example, the shock sensor write prohibition determination unit 411A can also determine, based on the write prohibition determination execution signal, whether the vibration or impact detected by the shock sensor 170 is greater than or less than a predetermined value. If it is determined that the vibration or impact is greater than the predetermined value, the shock sensor write prohibition determination unit 411A decides to prohibit (or stop) the writing (or writing operation) of the head HD1. Based on the determination result, the shock sensor write prohibition determination unit 411A generates and outputs a write prohibition determination signal for prohibiting (or stopping) the writing (or writing operation) of the head HD1.

[0079] On the other hand, if the vibration or impact is determined to be below a predetermined value, the impact sensor write prohibition determination unit 411A may also decide not to prohibit (or stop) the writing (or writing operation) of at least one head HD1 connected to the actuator AC1.

[0080] The impact sensor write prohibition determination unit 411B functions the same as the impact sensor write prohibition determination unit 411A in the control system B. Without specifically distinguishing between the impact sensor write prohibition determination units 411A and 411B, they are collectively referred to as the impact sensor write prohibition determination unit 411.

[0081] The HDC write inhibit determination unit 412A determines whether to inhibit (or stop) the writing (or writing operation) of the head HD1 based on the signal from the HDC60A, and outputs a control signal to the head amplifier IC30A based on the determination result. For example, if a write inhibit signal is received from the HDC60A to inhibit (or stop) the writing (or writing operation) of the head HD1, the HDC write inhibit determination unit 412A disables the write strobe (deassertes) and controls the head amplifier IC30A to inhibit (or stop) the writing (or writing operation) of the head HD1.

[0082] The HDC write prohibition determination unit 412B functions the same as the HDC write prohibition determination unit 412A in the control system B. Without specifically distinguishing between the HDC write prohibition determination units 412A and 412B, they are collectively referred to as the HDC write prohibition determination unit 412.

[0083] The MPU50A is a microprocessor unit (MPU) that outputs control signals to the driver IC20A based on signals from the HDC60A, etc. The MPU50A outputs control signals to cause the head HD1 to seek a predetermined location (e.g., target data sector TGT) on the disk DK1. Additionally, the MPU50A outputs signals for writing data to or reading data from a predetermined sector (data sector). For example, the MPU50A outputs signals for positioning the head HD1 at the target data sector TGT, writing data to or reading data from the predetermined sector (target data sector TGT).

[0084] MPU50B performs the same function as MPU50A in control system B. Unless otherwise specified, MPU50A and MPU50B are collectively referred to as MPU50.

[0085] The HDC60A is a hard disk controller, comprising a command control unit 610A, a servo control unit 620A, and a write action determination unit 630A. Each component of the HDC60A, such as the command control unit 610A, servo control unit 620A, and write action determination unit 630A, can be executed by firmware, software, or other programs. Alternatively, the HDC60A can also incorporate these components as hardware, such as circuitry. Furthermore, the HDC60A can be integrated into the MPU50A. The HDC60A controls the driver IC20A and the actuator block 100A.

[0086] The HDC60A controls the read / write processing of data on disk DK1 and controls the data transfer between host system 2 and R / W channel 40A. The HDC60A is electrically connected to, for example, R / W channel 40A, MPU50A, volatile memory 70A, buffer memory 80A, and non-volatile memory 90A.

[0087] HDC60B performs the same function as HDC60A in control system B. Unless otherwise specified, HDC60A and HDC60B are collectively referred to as HDC60.

[0088] The command control unit 610A acquires the status of actuator AC1 and selects commands stored in buffer memory 80A as command queue 1. The status of actuator AC1 includes the power mode indicating whether the actuator is operating, the number of waiting commands in command queue 1 scheduled to be processed in actuator AC1, and the Type_JerkSeek set as the operating parameter of actuator AC1. In addition, the status of actuator AC1 can also indicate the function of the control system (in this case, control system A) including actuator AC1, and the status of actuator block 100 (in this case, 100A).

[0089] The command control unit 610B obtains the status of the actuator AC2 and selects commands stored in the buffer memory 80B as command queue 2. The status of the actuator AC2 includes the power mode indicating whether the actuator is operating, the number of waiting commands in command queue 2 scheduled to be processed in the actuator AC2, and Type_JerkSeek set as the operating parameter of the actuator AC2.

[0090] Command control unit 610B functions the same as command control unit 610A in control system B. Unless otherwise specified, command control units 610A and 610B are collectively referred to as command control unit 610.

[0091] The actuator status confirmation unit 611A confirms the status of the actuator AC2 and determines the operating mode of the actuator AC1, which is the object of control of its own control system A.

[0092] The actuator status confirmation unit 611B confirms the status of actuator AC1 and determines the operating mode of actuator AC2, which is the object of control of its own control system B.

[0093] The actuator status verification unit 611B functions the same as the actuator status verification unit 611A in the control system B. Without specifically distinguishing between the actuator status verification units 611A and 611B, they are collectively referred to as the actuator status verification unit 611.

[0094] The actuator status communication unit 612A communicates with the actuator status communication unit 612B via the controller communication unit 190 to exchange data. For example, the actuator status communication unit 612A obtains the status of the actuator AC2, which is the control object of HDC60B.

[0095] The actuator status communication unit 612B functions similarly to the actuator status communication unit 612A in the control system B. For example, the actuator status communication unit 612B obtains the status of the actuator AC1, which is the object of control of the HDC60A. Without specifically distinguishing between the actuator status communication units 612A and 612B, they are collectively referred to as the actuator status communication unit 612.

[0096] The command selection unit 613A selects the command to be executed from the commands that are scheduled to be processed in the control system A and are stored in the command queue 1 in the buffer memory 80A.

[0097] Command selection unit 613B selects the command to be executed from the commands scheduled to be processed in control system B and stored in command queue 2 in buffer memory 80B. Command selection unit 613B performs the same function as command selection unit 613A in control system B. Unless otherwise specified, command selection units 613A and 613B are collectively referred to as command selection unit 613.

[0098] The rearrangement list selection unit 6131A selects a rearrangement list, which is used to determine the next command to be processed from the commands stored in the command queue 1.

[0099] The rearrangement list selection unit 6131B selects a rearrangement list, which is used to determine the next command to be processed from the commands stored in the command queue 2. The rearrangement list selection unit 6131B functions the same as the rearrangement list selection unit 6131A in the control system B. Unless otherwise specified, the rearrangement list selection units 6131A and 6131B are collectively referred to as the rearrangement list selection unit 6131.

[0100] The anticipatory seek time limit command selection unit 6132A checks whether there is a command in the command queue 1 whose seek time (anticipatory time) is longer than the end time (anticipatory moment) of the command being processed by actuator AC2. Based on the check result, the anticipatory seek time limit command selection unit 6132A selects, for example, JIT seek as the action characteristic of seek acceleration. JIT seek will be described later.

[0101] The anticipatory seek time limit command selection unit 6132B checks whether there are commands in the command queue 2 whose seek time (anticipatory time) is longer than the end time (anticipatory moment) of the command being processed by actuator AC1. Based on the check result, the anticipatory seek time limit command selection unit 6132B selects, for example, JIT seek as the action characteristic for seek acceleration. The anticipatory seek time limit command selection unit 6132B functions similarly to the anticipatory seek time limit command selection unit 6132A in the control system B. Without specifically distinguishing between the anticipatory seek time limit command selection units 6132A and 6132B, they are collectively referred to as the anticipatory seek time limit command selection unit 6132.

[0102] The servo control unit 620A controls the position of the head HD1. In other words, the servo control unit 620A controls the access of the head HD1 to a predetermined area (e.g., the target data sector TGT) of the disk DK1. The servo control unit 620A includes a tracking control unit 621A and a seek control unit 622A.

[0103] The servo control unit 620B performs the same function as the servo control unit 620A in the control system B. Without specifically distinguishing between the servo control units 620A and 620B, they are collectively referred to as the servo control unit 620.

[0104] The tracking control unit 621A controls the head HD1 to track a predetermined track on disk DK1. Sometimes, the tracking of the head HD1 to a predetermined track on disk DK1 is simply referred to as tracking. Tracking includes tracing a predetermined path (e.g., a predetermined track) when writing data to disk DK1 and tracing a predetermined path (e.g., a predetermined track) when reading data from disk DK1.

[0105] The tracking control unit 621B performs the same function as the tracking control unit 621A in the control system B. Without specifically distinguishing between the tracking control units 621A and 621B, they are collectively referred to as the tracking control unit 621.

[0106] The seek control unit 622A performs seek control on the head HD1 towards the target track in disk DK1. The seek control unit 622A determines the seek trajectory for the head HD1. Furthermore, the seek control unit 622A switches the seek control operation mode according to Type_JerkSeek, which is the seek operation mode set by the actuator status confirmation unit 611A.

[0107] The seek control unit 622B performs the same function as the seek control unit 622A in the control system B. Unless otherwise specified, the seek control units 622A and 622B are collectively referred to as the seek control unit 622.

[0108] When Type_JerkSeek is set to 1 by the actuator status confirmation unit 611A, the low Jerk seek control unit 6221A performs seek control based on the low Jerk operation characteristics.

[0109] When Type_JerkSeek is set to 1 by the actuator status confirmation unit 611B, the low-Jerk seek control unit 6221B performs seek control based on the low-Jerk operating characteristics. Unless otherwise specified, the low-Jerk seek control units 6221A and 6221B are collectively referred to as the low-Jerk seek control unit 6221.

[0110] When the Type_JerkSeek setting is 2 as set by the actuator status confirmation unit 611A, the high-Jerk seek control unit 6222A performs seek control based on the high-Jerk motion characteristics.

[0111] When the Type_JerkSeek setting is 2 as set by the actuator status confirmation unit 611B, the low Jerk seek control unit 6221B performs seek control based on the high Jerk operating characteristics. Without specifically distinguishing between the high Jerk seek control units 6222A and 6222B, they are collectively referred to as the high Jerk seek control unit 6222.

[0112] When the Type_JerkSeek setting is 0 as set by the actuator status confirmation unit 611A, the JIT seek control unit 6223A performs seek control based on the JIT seek operation characteristics.

[0113] When the Type_JerkSeek setting, as set by the actuator status confirmation unit 611B, is 0, the JIT seek control unit 6223B performs seek control based on the JIT seek operation characteristics. Unless otherwise specified, the JIT seek control units 6223A and 6223B are collectively referred to as JIT seek control unit 6223.

[0114] The write action determination unit 630A determines the write action of the head HD1. The write action determination unit 630A includes a position write action determination unit 631A and a speed write action determination unit 632A. The position write action determination unit 631A determines the write action of the head HD1 based on its position. The speed write action determination unit 632A determines the write action of the head HD1 based on its speed.

[0115] The write action determination unit 630B functions the same as the write action determination unit 630A in the control system B. Without specifically distinguishing between the write action determination unit 630A and the write action determination unit 630B, they are collectively referred to as the write action determination unit 630.

[0116] Volatile memory 70A and volatile memory 70B (collectively referred to as volatile memory 70 unless otherwise distinguished) are semiconductor memories whose stored data is lost when the power supply is disconnected. Volatile memory 70 stores data required for processing in various parts of disk drive 1. Volatile memory 70 is, for example, DRAM (Dynamic Random Access Memory) or SDRAM (Synchronous Dynamic Random Access Memory). Volatile memory 70A and volatile memory 70B can also be used as working memory in the processing of control system A and control system B, respectively.

[0117] Buffer memory 80A and buffer memory 80B (collectively referred to as buffer memory 80 unless otherwise distinguished) are semiconductor memories that temporarily store data transmitted between disk drive 1 and host system 2. For example, buffer memory 80 temporarily stores commands received by disk drive 1 from host system 2. Furthermore, buffer memory 80 may be physically integrated with volatile memory 70. Additionally, buffer memory 80A may be physically integrated with buffer memory 80B. Buffer memory 80 may be, for example, DRAM, SRAM (Static Random Access Memory), SDRAM, FeRAM (Ferroelectric Random Access Memory), or MRAM (Magnetoresistive Random Access Memory).

[0118] The buffer memory 80A is the destination for storing command queue 1 (waiting for command 1) for DK1 of control system A.

[0119] The buffer memory 80B is the destination for storing command queue 2 (waiting command 2) for DK2 of control system B.

[0120] Non-volatile memory 90A and non-volatile memory 90B (collectively referred to as non-volatile memory 90 unless otherwise distinguished) are semiconductor memories that retain stored data even when the power supply is interrupted. Non-volatile memory 90 is, for example, a NOR or NAND type flash ROM (Flash Read Only Memory). Rearrangement lists and similar data can also be stored in non-volatile memory 90.

[0121] Shock sensor 170 detects vibrations and / or shocks applied from the outside to the disk drive 1 or the housing HS of the disk drive 1's HDA10, etc. Upon detecting vibration and / or shock, shock sensor 170 outputs a signal indicating that vibration and / or shock has been detected (hereinafter sometimes referred to as a vibration / shock detection signal). Shock sensor 170 is electrically connected, for example, to write inhibit detector 180. Upon detecting vibration and / or shock, shock sensor 170 outputs a vibration / shock detection signal to write inhibit detector 180.

[0122] The write prohibition detector 180 outputs a signal (sometimes called a write prohibition determination execution signal) to determine whether to perform a write (or write operation) prohibition. Upon receiving a vibration / shock detection signal, the write prohibition detector 180 outputs the write prohibition determination execution signal. The write prohibition detector 180 is, for example, electrically connected to the R / W channel 40. Upon receiving a vibration / shock detection signal, the write prohibition detector 180 outputs the write prohibition determination execution signal to the R / W channel 40.

[0123] The controller communication unit 190 controls the information transmission between multiple control systems. The controller communication unit 190 is electrically connected to, for example, control system A and control system B. The controller communication unit 190 includes an actuator status communication unit 191. The actuator status communication unit 191 exchanges actuator states between HDC60A and HDC60B.

[0124] Furthermore, not intending to Figure 1 The configuration shown specifically limits the number and composition of each function. Furthermore, while the configurations of control systems A and B are shown in this embodiment, no particular limitation is intended.

[0125] Figure 3 This is a block diagram illustrating the functional configuration of the servo control unit of the disk device according to this embodiment.

[0126] In the servo control unit 620, by setting any one of the target position 6201, target velocity 6202, and target acceleration 6203, the current indication value determination unit 6204 determines the current and voltage applied to the actuators (VCM, MA, etc.). The current application unit 6205 moves the actuator AC by applying current to the VCM, and demodulates the position by reading the position information of the servo mode on the disk DK through the playback head. Alternatively, the structure can be as follows: instead of directly using the position information read from the servo mode on the disk DK, the difference between the estimated position (based on position, velocity, and acceleration (current)) and the demodulated position is input to the status estimation unit 6207, and the estimated position and estimated velocity are updated in each servo sample. Furthermore, the device (plant) 6206 represents the function of the voice coil motor (VCM) and multi-actuator (MA) used in a typical HDD.

[0127] Figure 4 This diagram illustrates an example of the effect of the actuator of the disk device according to this embodiment on other actuators when it starts up.

[0128] In a multi-actuator disk drive, when any actuator AC performs a seek operation, it applies vibrations to the other actuator ACs. The actuator AC that performs the seek operation is called the aggressor, and the actuator AC that receives the vibrations from the aggressor is called the victim.

[0129] Feature 1011 represents an example of the aggressor's actions under high Jerk seek control. Feature TC11 represents an example of the time characteristics of the VCM acceleration of the aggressor's head tilt HD. Feature TC12 represents the Jerk (jerk) relative to feature TC11, indicating the time variation of acceleration a (da / dt). Feature TC13 represents an example of the time characteristics of the head position error of the victim's head tilt HD when the aggressor's head tilt HD performs actions as described in feature TC11.

[0130] Feature 1021 represents an example of the aggressor's actions under low Jerk seek control. Feature TC21 represents an example of the time characteristics of the VCM acceleration of the aggressor's head tilt HD. Feature TC22 represents the Jerk (jerk) relative to feature TC21, indicating the time variation of acceleration a (da / dt). Feature TC23 represents an example of the time characteristics of the head position error of the victim's head tilt HD when the aggressor's head tilt HD performs actions as described in feature TC21.

[0131] TH11 and TH12 represent the threshold values ​​for features TC12 and TC22, respectively. TH11 is set to th... jerk TH12 is set to -th jerkIn characteristic 1011, since it is a characteristic for high Jerk seek control, characteristic TC12 shows the cases where the threshold TH11 is exceeded and the threshold TH12 is exceeded. Furthermore, in characteristic 1021, since it is a characteristic for low Jerk seek control, characteristic TC22 is controlled so that the threshold TH11 is not exceeded and the threshold TH12 is not exceeded.

[0132] TH21 and TH22 represent the threshold values ​​for features TC13 and TC23, respectively. TH21 is set to th... PES TH22 is set to -th PES In characteristic 1011, since it is a characteristic for high Jerk seek control, characteristic TC13 shows a situation where the head position error exceeds the threshold TH21 and falls below TH22, indicating a large head position error on the victim side. Conversely, in characteristic 1021, since it is a characteristic for low Jerk seek control, characteristic TC23 is controlled to prevent exceeding the threshold TH21 and falling below TH22, thus minimizing the head position error on the victim side.

[0133] As described above, when writing and reading data sectors are performed on the victim's side, write operations may be disabled or read operations may fail due to misalignment of the head position. Therefore, this embodiment may also include a function to suppress the effects of vibration interference generated between the actuators AC.

[0134] More specifically, when the aggressor's head HD performs a pathfinding action, disturbance information such as positional disturbances that affect the victim's head HD is estimated, and the estimated disturbance information is reflected in the action of the miniature actuator MA of the victim's head HD to control the victim's head HD.

[0135] Figure 5 This is a diagram showing the data flow in the disk device according to this embodiment, specifically the function of suppressing vibration interference between actuators. This function will be described as a function of the servo control unit 620.

[0136] Here, the actuator AC1 of control system A will be described as the aggressor, and the actuator AC2 of control system B will be described as the victim. Furthermore, in this embodiment, both control systems A and B have the same function; therefore, the aggressor and victim can be interchanged by switching them.

[0137] When the servo control unit 620A performs a seek operation, the controller 6241A determines the VCM operation quantity (control current) of the intruder. This VCM operation quantity is then input to the digital filter 6244A (transfer characteristic F) for mutual interference between the actuators AC and AC.xact This allows for the estimation of positional perturbations that affect the effective reproduction head HD2 on the victim's side. Transmission characteristics F xact This indicates the transfer characteristics of a digital filter.

[0138] The transmission characteristic F is determined by the distinguishing radius positions of effective heads HD1 and disk DK1 on the aggressor side and the distinguishing radius positions of effective heads HD2 and disk DK2 on the victim side. xact It can change, therefore, like the digital filters 6244A and 6244B, it can have a function in its respective configuration to generate the transfer characteristic F. xact The parameters.

[0139] Specifically, the control current of the VCM is input from the controller 6241A to the VCM 6243A, and simultaneously input to the digital filter 6244A. The output current of the digital filter 6244A (such as...) Figure 4 A current with a waveform similar to TC13 is input to the servo control unit 620B of the control system B. In the servo control unit 620B, the control current output by the controller 6241B is combined with the output current of the digital filter 6244A and input to the micro-actuator MA6242B. Through the output current of the digital filter 6244A, the servo control unit 620B performs an addition operation on the micro-actuator MA6242B of the effective playback head on the victim side, causing the effective playback head on the victim side to operate, thereby compensating for the estimated position disturbance affecting the effective playback head HD on the victim side. This reduces the position error PES of the effective playback head HD on the victim side relative to the target track.

[0140] When writing and reading data sectors of disk DK2 are performed on the victim's side, if the head position error exceeds the head position error threshold ±th... PES When the range is defined, write prohibition is performed during write operations, and read operations are stopped if a read error is detected.

[0141] In addition, in this embodiment, such as Figure 4 Similar to the timing characteristics of TC22, the action of the aggressor-side actuator AC1 during seek is low Jerk seek, limiting Jerk to ±th. jerk Within this range. Therefore, the position error of the tracking head HD2 of the actuator AC2 on the victim's side can be reduced to within ±th. PES The data sector can be read and written within the specified range.

[0142] Figure 6 This is a diagram illustrating the rearrangement of the table involved in this implementation.

[0143] Table RTD1 represents the rearrangement list used for low Jerk seek, and table RTD2 represents the rearrangement list used for high Jerk seek. Data STD1 represents the seek time during low Jerk seek control, and data STD2 represents the seek time during high Jerk seek control. The rearrangement lists (RTD1, RTD2) and seek time characteristics (STD1, STD2) can also be stored in non-volatile memory 90 and expanded to volatile memory 70 when disk device 1 is started. Alternatively, they can also be stored in volatile memory 70 and buffer memory 80 when selected by the rearrangement list selection unit 6131.

[0144] In this embodiment, the actuator AC is configured to at least be capable of performing actions having Figure 4 The characteristics of TC22 are shown to be low Jerk characteristics controlled within the ±thjerk range, and Figure 4 Characteristic TC12 illustrates a high Jerk characteristic seek control that exceeds at least the ±thjerk range. In this control (from a seek time perspective), it is desirable, for example, when controlling the head HD1 via the aggressor's actuator AC1, whether using a low Jerk characteristic-based seek control or a high Jerk characteristic-based seek control, to ensure that the peak value of the VCM acceleration is consistent, as shown by characteristics TC11 and TC21. Furthermore, in the seek control of characteristics TC12 and TC22, at least one of the VCM current peak and acceleration peak may be the same or different.

[0145] Furthermore, when actuator AC2 is not performing a seek operation, actuator AC1 performs a high-jerk seek operation exceeding the ±thjerk range. This shortens the seek time of actuator AC1, allowing for command access in a shorter time and improving command access performance.

[0146] However, on the other hand, when it becomes controlled within ±th jerk When the range is low Jerk, such as Figure 6 As shown in STD1, sometimes the seek time is long, and the command access performance (the number of commands that can be accessed per unit time) will decrease.

[0147] Figure 7 This is a timing diagram illustrating the seek operation performed by the disk device according to this embodiment, showing an example of the overall flow from receiving a command sent from the host system 2 to the seek operation. Here, the command sent from the host system 2 is assumed to be a command to disk DK1 processed by the control system A.

[0148] The host system 2 sends a data read or write request (called a command) Cmd21 (SC21) to the disk device 1. When the disk device 1 receives the command Cmd21 from the host system 2, it saves it to the empty portion of the command queue 1 (CQ1-1) on the buffer memory 80A of the disk device 1. Furthermore, Figure 7 CQ1-1 to 1-5 in the table represent the same command queue 1, showing the time transition of the state.

[0149] Before SC21, the command control unit 610A selects the command Cmd11 executed in SC101 and issues a seek request to the servo control unit 620A for accessing data on DK1 specified by command Cmd11 (SC102). Based on the time (expected time) at which the header HD1 arrives at the target data sector TGT for the data specified by command request Cmd11, the servo control unit 620A selects a seek trajectory mode within the expected time from seek trajectories corresponding to the disk radius distance traveled by the header HD1 towards the target data sector TGT (SC201). This enables writing and reading from the target data sector TGT.

[0150] The servo control unit 620A performs seek control of the head HD1 on the target data sector TGT using the seek trajectory of the selected mode (SC202). After the head HD1 moves to the track containing the target data sector TGT through seek control, it issues a seek completion notification to the command control unit 610A (SC203). The command control unit 610A waits for the disk DK1 to rotate, and when the target data sector TGT is detected (SC104), it controls the head amplifier IC30A to perform reading or writing of the data specified by the read head RH1 or the write head WH1 (SC105).

[0151] By following the steps above, data can be read or written to disk DK1 via Cmd11. Control system A, for example, HDC60A, notifies host system 2 that Cmd11 has completed this process (SC106).

[0152] In SC201 to SC106, while the head HD1 and others are processing Cmd11, the command control unit 610A selects the command to be executed next from the commands stored in CQ1-2 (SC103). In SC103, the status of the other actuator AC2 is checked, and the Jerk setting and command Cmd12 are selected based on the status of actuator AC2. When the processing of Cmd11 is completed in SC106, the processing of Cmd12 is performed in the same way as in SC201 to SC106 (SC107, SC204, SC205). In addition, regarding the new command Cmd22 from the host system 2, it is also stored in the blank part of the command queue 1 (CQ1-4), and in the same way as in SC103, the command control unit 610A selects the command to be executed next from the commands stored in CQ1-5 (SC108). Subsequently, disk device 1 repeatedly performs the same steps based on the request commands from host system 2.

[0153] Figure 8 This is a flowchart illustrating the command selection process for the disk device according to this embodiment, where the command control unit 610A is used to select commands. Figure 7 The flowchart illustrates the selection of commands from command queue 1 in the command selection (rearrangement) of SC101, SC103, and SC108. The flowchart is explained as part of the processing in actuator AC1 and the control system A that controls it.

[0154] In the command control unit 610A, the actuator status confirmation unit 611A confirms the status of actuator AC2 in the control system B (step S1). The rearrangement list selection unit 6131A selects a rearrangement list based on the status of actuator AC2 confirmed in step S1 (step S2). The command selection unit 613A selects a command from the command queue 1 based on the rearrangement list selected in step 2 (step S3). In step S3, for example, the command with the shortest access time (seek time + rotation wait time) in the command group is selected.

[0155] Figure 9 This is a flowchart illustrating the type of Jerk seek control set for the disk device involved in this implementation, equivalent to... Figure 8 The processing of S1.

[0156] The actuator status communication unit 612A obtains the power mode and the number of hold commands held in command queue 2 as status information of actuator AC2 from the actuator status communication unit 612B of the control system B (step S101). The actuator status confirmation unit 611A confirms whether the power mode obtained in step S101 is active (step S102). If the power mode of actuator AC2 in control system B is not active (step S102: No), the actuator status confirmation unit 611A sets its own parameter Type_JerkSeek to 2, indicating high Jerk seek control (step S107). The case where the power mode is not active means, for example, the power mode is standby, idle, or sleep. That is, when the power mode of actuator AC2 is not active, HDC60A performs seek control of actuator AC1 with high Jerk.

[0157] On the other hand, when the power mode of actuator AC2 in control system B is activated (step S102: Yes), actuator status confirmation unit 611A confirms the number of held commands in command queue 2 obtained in step S101 (step S103).

[0158] When the number of commands in command queue 2 is 0 (step S103: No), the actuator status confirmation unit 611A sets its own parameter Type_JerkSeek to 2, indicating high Jerk seek control (step S107). When the number of commands in actuator AC2 is 1 or more (step S103: Yes), the actuator status confirmation unit 611A confirms the Type_JerkSeek set in actuator AC2 (step S104).

[0159] When Type_JerkSeek = 2 (step S104: Yes), the actuator status confirmation unit 611A sets its own parameter Type_JerkSeek to 0, indicating low Jerk seek control (step S105). Conversely, when Type_JerkSeek is other than 2 (step S104: No), the actuator status confirmation unit 611A sets its own parameter Type_JerkSeek to 1, indicating high Jerk seek control (step S106).

[0160] The above steps for setting Type_JerkSeek can be summarized as follows: When the power mode of actuator AC2 on the other side (control system B) is standby or idle, set high Jerk seek control (Type_JerkSeek = 2). When the power mode of actuator AC2 on the other side is active and the number of commands in command queue 2 related to actuator AC2 is 0, set high Jerk seek control (Type_JerkSeek = 2). When the power mode of actuator AC2 on the other side is active, the number of commands in command queue 2 related to actuator AC2 is 1 or more, and Type_JerkSeek of actuator AC2 is 2, set low Jerk seek control (Type_JerkSeek = 0). If the power mode of the other actuator AC2 is active, the number of commands in the command queue 2 involved by the other actuator AC2 is 1 or more, and the Type_JerkSeek of the other actuator AC2 is not 2, set low Jerk seek control (Type_JerkSeek=1).

[0161] Figure 10 This is a flowchart illustrating the type of disk device setting reorder list involved in this implementation, equivalent to... Figure 8 The processing of S2.

[0162] Rearrange the list selection section 6131A for... Figure 9 In the process, the actuator status confirmation unit 611A confirms the Type_JerkSeek set (step S201). In the case of a low Jerk seek where Type_JerkSeek is 0 or 1 (step S201, '0, 1'), the rearrangement list selection unit 6131A retrieves the rearrangement list for low Jerk seek from the volatile memory 70A and sets it in the servo control unit 620A (step S202). In the case of a high Jerk seek where Type_JerkSeek is 2 (step S201, '2'), the rearrangement list for high Jerk seek is retrieved from the volatile memory 70A and set in the servo control unit 620A (step S203).

[0163] Figure 11 This is a flowchart illustrating the command selection process for the disk device involved in this implementation, equivalent to... Figure 8 The processing of S3.

[0164] Command selection unit 613A for in Figure 9In the process, the actuator status confirmation unit 611A confirms the Type_JerkSeek set (step S301). If the Type_JerkSeek set by the actuator status confirmation unit 611A is not 0 (step S301: No), the command selection unit 613A performs normal command selection processing (step S304). That is, in step S304, the command selection unit 613A selects a command with a short access time from the command queue 1 of the control system A as usual. The servo control unit 620A processes the command selected in step S304. More specifically, when Type_JerkSeek = 1, the servo control unit 620A performs seek control of the head HD1 in low Jerk seek mode, and when Type_JerkSeek = 2, the servo control unit 6221A performs seek control of the head HD1 in high Jerk seek mode.

[0165] In step S301, when Type_JerkSeek is 0 (0 indicates low Jerk seek control. However, the other actuator AC2 is set to high Jerk seek control) (step S301: Yes), the command selection unit 613A confirms the end time of the command currently being executed on the actuator AC2 side and the access time of each command in the command queue 1 of the control system A (step S302).

[0166] In step S302, if there are no commands in command queue 1 that require a seek time longer than the end time of commands executed on the other actuator AC2 side (step S302: No), the command selection unit 613A performs normal command selection processing (step S304). That is, it selects commands with shorter access times from the command group in command queue 1 for control system A as usual (step S304). In step S304, it is desired that the servo control unit 620A perform JIT seek for the time after adding one rotation. This reduces power consumption caused by seek.

[0167] In step S302, if command queue 1 contains commands that require a longer seek time than the command end time on the other actuator AC2 side (step S302: Yes), command selection unit 613A selects a command with a shorter access time from the commands in command queue 1 that require a longer seek time than the command end time on the other actuator AC2 side (step S303). In step S303, it is then desired that servo control unit 620A perform seek using JIT seek. This reduces power consumption caused by seek.

[0168] When summarizing the above steps, they can be recorded as follows.

[0169] During the high-jerk seek operation of actuator AC2 in control system B, there is a problem that control system A cannot access data due to the vibration caused by control system B. To avoid this problem, control system A controls the head HD1 to access data (data read / write processing) after the high-jerk seek operation implemented by control system B ends. For example, control system A can also perform read / write processing through head HD1 after moving head HD1 to the target track by seek control and after the waiting disk DK1 has rotated once, during the processing of the command selected from command queue 1 through normal command selection processing (equivalent to step S304). Furthermore, control system A can also perform data read / write processing after waiting for the further rotation time of disk DK1, even if the high-jerk seek operation implemented by control system B does not end after the waiting disk DK1 has rotated once.

[0170] In addition, regarding steps S303 and S304, as a method for selecting commands with shorter access times from the command queue 1 of the control system A, the following method can also be adopted: not only considering the step after step 1, but also the step after step 2, step 3, etc., to select the command after step 1 where the time in the route becomes shorter.

[0171] The command selection unit 613A executes the command selected through the above process. The servo control unit 620A controls the driver IC20, actuator AC1, etc. based on the command selected by the command selection unit 613A.

[0172] Figure 12 This is a flowchart illustrating the seek control process performed by the disk device according to this embodiment, showing that... Figure 11 The processing of the selected command in step S303. Seek control refers to the control that moves the head HD to the track (target track) of the target data sector specified by the command.

[0173] The JIT seek control unit 6223A of the servo control unit 620A selects a seek trajectory mode sufficient to handle the data sector (target data sector TGT) (step S11). In the selection of seek trajectory mode, for example, if the access performance is the same, it is better to have low power consumption caused by seek operation. In step S11, the JIT seek control unit 6223A selects, for example, the seek trajectory mode with the lowest power consumption (also known as JIT seek).

[0174] Figure 13A This is a diagram showing the timing of seek control performed by the two control systems of the disk device involved in this embodiment.

[0175] Figure 13A(a) represents the state SA1 of control system A and the state SB1 of control system B on the same time axis, which is equivalent to Figure 11 The states of control systems A and B are defined as "Yes" in S301. State SB1 of control system B shows that control system B is processing a command (let's call it command B) for the target data sector TGT-B. At time T1, control system B is performing seek control with high Jerk seek. State SA1 of control system A shows that while control system B is performing seek control with high Jerk seek, control system A can perform low Jerk-based seek control for actuator AC1, but cannot perform read / write control for the target data sector. Control system B ends high Jerk seek control at time T3, making read / write control of the target data sector possible through control system A.

[0176] Figure 13A (b) represents the state SA2 of control system A and the state SB2 of control system B on the same time axis, which is equivalent to Figure 11 The state of control systems A and B is "Yes" in step S302. At time T1 when control system A is in state SA2, command selection unit 613A compares the expected end time T5 of seek control of the target data sector TGT-A for the command to be processed by control system A (let's call it command A) with the expected completion time T3 of seek control based on high Jerk seek control of control system B. Command selection unit 613A of control system A selects JIT seek for seek control of the target data sector TGT-A.

[0177] Figure 13A (c) represents the state SA3 of control system A and the state SB3 of control system B on the same time axis, which is equivalent to... Figure 11 After step S302 is "No", in S304, a command is selected, and the seek operation is performed using the waveform of the corresponding JIT seek mode. This is the state of control systems A and B. At time T1, in state SA3 of control system A, the command selection unit 613A compares the expected end time T2 of the seek control to target data sector TGT-A for the command to be processed by control system A (let's call it command A) with the expected completion time T3 of the seek control based on high-Jerk seek control of control system B. The command selection unit 613A of control system A selects JIT seek for the seek control to target data sector TGT-A. However, in... Figure 13AIn case (c), the start time T2 of the read / write control of the target data sector TGT-A based on control system A overlaps with the high Jerk seek control of control system B, which greatly increases the possibility of errors in the read / write control of TGT-A. To avoid this error, control system A could, for example, wait for the high Jerk seek control of control system B to finish before performing the read / write control of the target data sector TGT-A.

[0178] Figure 13A (d) is equivalent to in Figure 13A In case (c), the states of control systems A and B are as follows: control system A waits for control system B's high-Jerk seek control to finish before performing read / write control on the target data sector TGT-A. Figure 13A Example (d) illustrates the case where control system A waits for one full disk rotation. Control system A begins read / write control of target data sector TGT-A at time T6, after one full disk rotation has elapsed since the start of read / write control from the initial start time T2. Control system A selects a JIT seek that ends the seek control to target data sector TGT-A at time T6. Thus, control system A is able to execute read / write control of target data sector TGT-A without being affected by the high Jerk seek control of control system B.

[0179] Figure 13B This is a diagram showing the relationship between the operating states of the two control systems of the disk device involved in this embodiment.

[0180] Tracking control is used to position the head HD on the target track, and can also be considered as the state when no seek is performed (including high Jerk seek, low Jerk seek, etc.). For example, in Figure 13A The display shows the time after the TGT (TGT-B, TGT-A) data read / write time, the time before the seek begins, and the time after the TGT (TGT-B, TGT-A) data read / write time, indicating the "tracking control" state. It can also be used in... Figure 9 The action parameter Type_JerkSeek set in the configuration can be set to -1 for actuator block 100 or actuator AC during tracking control. Alternatively, other values ​​can be assigned to Type_JerkSeek for the state of head HD reading and writing data.

[0181] State STT1 indicates the permitted operating state for the other actuator block (apparently actuator block 100A) when one actuator block (apparently actuator block 100B) is performing a seek operation with high Jerk seek control. That is, when actuator block 100B is performing a seek operation with high Jerk seek control, actuator block 100A is configured to perform tracking control, disable head HD1 data reading / writing, enable low Jerk seek control, and disable high Jerk seek control, respectively, to implement control.

[0182] Status STT2 indicates that the actuator block 100B is not able to perform tracking control, head HD1 data reading and writing, low Jerk seek control, or high Jerk seek control when performing seek control with high Jerk seek control.

[0183] Figure 14 This is a diagram illustrating an example of the relationship between acceleration and position in seek control of a disk device according to this embodiment.

[0184] TC100 represents an example of the time characteristic of the seek acceleration of head HD1 (time versus seek acceleration), TC200 represents an example of the time characteristic of the seek position of head HD1 (time versus seek position), characteristics TC101 and TC201 represent the characteristics under the fastest seek condition, and characteristics TC102 and TC202 represent the characteristics under the JIT seek condition. The target data sector TGT201 is set to be one sector on one track of disk DK1.

[0185] Fastest seek is the fastest seek mode that allows the head HD1 to reach the target data sector TGT201, for example... Figure 4 The TC11 shown is a case where the head HD1 is moved by high Jerk seek control.

[0186] Just-in-Time (JIT) seek refers to seek control using a seek trajectory optimized to reduce power consumption of the disk device 1 due to seek control. More specifically, JIT seek refers to a seek trajectory optimized to reduce power consumption, etc., in a seek mode where seek is completed immediately before the target data sector.

[0187] In accordance with the actuator state verification unit 611A Figure 9 In the process, when the fastest seek mode characteristic TC101 in the Type_JerkSeek condition is used for seek, the track with the target data sector TGT (target track) is reached ahead of (delta)Tseek time (REF202). Figure 12 In step S11, a Seek mode with a longer Seek time corresponding to the (delta)T seek time is selected, i.e., JIT seek. For example, in Figure 14 In the JIT seek characteristic TC102, the acceleration level in the acceleration and deceleration intervals is reduced (the voltage level applied to the VCM is reduced) compared to the fastest seek mode characteristic TC101. In order to adjust the seek distance to the target track, the length of the zero acceleration interval (which is near zero acceleration, also known as the constant speed interval) is adjusted.

[0188] By reducing the acceleration level, the power consumption P of the disk device 1 is reduced. The power consumption P is determined by the current I flowing in the VCM. vcm The current I flowing through the VCM is determined by the circuit resistance R, including the VCM, when the acceleration level is reduced. vcm As the power consumption P decreases, the power consumption P also decreases.

[0189] P = R × I vcm 2

[0190] Furthermore, the above examples illustrate how power consumption can be reduced by decreasing the acceleration level. However, power consumption can also be reduced by shortening the acceleration and deceleration intervals and adjusting the length of the zero-acceleration interval (which is near zero acceleration, also known as the constant speed interval).

[0191] In adjusting the seek control of characteristic TC102, any combination of time or acceleration level in the acceleration range, time or acceleration level in the deceleration range, and time in the constant speed range of actuator AC1 can be used for adjustment.

[0192] The seeker control unit 622A is in Figure 12 In step S11, after selecting the seek trajectory mode such as acceleration characteristic T101, seek control is executed according to the selected seek trajectory mode (seek control setting). More specifically, the seek control unit 622A controls the actuator AC1 according to its own parameter Type_JerkSeek set by the actuator state confirmation unit 611A. For example, when Type_JerkSeek = 0, the low Jerk seek control unit 6221A or the JIT seek control unit 6223A controls the actuator AC1.

[0193] By taking the above steps into account the state of actuator AC2, actuator AC1 can be activated, which can shorten the seek time, allow access to commands in a short time, and improve command access performance.

[0194] (Modified Example)

[0195] The embodiment shows an example of a disk having two control systems that control two disks (DK) and actuators (AC), but there can be more than two disks (DK), actuators (AC), and control systems. For example, the spindle 12 may have two or more disks (DK), each with an actuator block 100. Furthermore, there may be two or more spindles 12, each with two or more disks (DK) and actuator blocks 100 controlling those disks, and there may be two or more control systems. In the case where the disk drive 1 has two or more actuators (AC) (or actuator blocks 100), for example... Figure 9 Steps S102, S103, S104 Figure 13B The “other actuators AC (or actuator block 100)” in the above can also be set to multiple actuators AC (or actuator block 100) other than the actuator AC (or actuator block 100) of interest.

[0196] Figure 15 This is a schematic diagram of the disk device involved in the variation example.

[0197] In this variant, the disk device is identical to HDA10-2, except that HDA10 is set to HDA10-2. Figure 1 The disk device 1 is the same. The HDA10-2 has two actuator blocks 100 mounted in the horizontal direction of a spindle (SPM) 12 (assuming the HDA10 is vertical). Figure 2 (a) is the same example. With Figure 1 Unlike the HDA10, in the HDA10-2, actuator blocks 100A and 100B are respectively installed on the two coaxial BR2 used in the VCM.

[0198] In this modified disk device, through two control systems A and B (and...) Figure 1 Similarly, the actuator blocks 100A and 100B are controlled in the same way as the control shown in the various embodiments, enabling command access that takes into account the effects caused by multiple actuators.

[0199] The features shown in this embodiment are extracted as follows.

[0200] (A-1) A disk drive having at least two or more independently drivable actuator blocks 100 for low Jerk seek control with limited Jerk, characterized in that, in the disk drive 1, when actuator block 100B is not accessing a data sector of the disk, actuator block 100A accesses a data sector of the disk through high Jerk seek control.

[0201] (A-2) According to the disk device described in (A-1), the state in which the actuator block 100B is not accessing the data sector of the disk is either when the power mode of the actuator block 100B is not activated or when the command queue 2, which holds commands as access requests from the host system 2 to the disk, is empty.

[0202] (A-3) The disk device according to (A-1) is characterized in that a low Jerk seek control parameter and a high Jerk seek control parameter are maintained in the non-volatile memory, the low Jerk seek control parameter being used to perform the low Jerk seek control that limits Jerk, and the high Jerk seek control parameter being used to perform the high Jerk seek control.

[0203] (A-4) The disk device according to (A-3) is characterized in that, at startup, the low Jerk seek control parameters and the high Jerk seek control parameters stored in the non-volatile memory are expanded in the volatile memory.

[0204] (A-5) The disk device according to (A-1) is characterized in that, in the low Jerk seek control which limits Jerk and the high Jerk seek control, each has a correspondence table of seek time relative to seek distance for implementing command rearrangement.

[0205] (A-6) The disk device according to (A-5) is characterized in that the correspondence table is switched when switching from the seek control with low Jerk that limits Jerk to the seek control with high Jerk.

[0206] (A-7) The disk device according to any one of (A-1) to (A-6) is characterized in that, during the period when the actuator block 100B is performing the high-jerk seek control, when the actuator block 100A is performing seek control, if there are more than one command in the command queue 1 that can access the data sector controlled by the actuator block 100A at a time later than the seek end time of the actuator block 100B, the command to be accessed next is selected from the commands in the command queue 1.

[0207] (A-8) The disk device according to any one of (A-1) to (A-7) is characterized in that, during the period when actuator block 100B is performing the high-jerk seek control, when actuator block 100A is performing seek control, if there is no command in command queue 1 that can access the data sector at a time later than the seek end time of actuator block 100B, actuator block 100A selects the next access command from command queue 1, and inserts one revolution of disk time after the seek end time of actuator block 100B to enable access.

[0208] (A-9) The disk device according to any one of (A-1) to (A-8) is characterized in that, after the target data sector is determined in the seek control of the actuator block 100A, the seek control of the actuator block 100A is adjusted so that the seek of the actuator block 100A ends just before the moment when the target data sector is about to be accessed.

[0209] (A-10) The disk device according to (A-9) is characterized in that, in the adjustment of the seek control of the actuator block 100, the adjustment is made by any combination of the time or acceleration level in the acceleration interval, the time or acceleration level in the deceleration interval, and the time in the constant speed interval of the actuator AC.

[0210] (A-11) The disk device according to any one of (A-1) to (A-10) is characterized in that, in the seek control with high Jerk and the seek control with low Jerk performed by the actuator block 100, the peak level of either the acceleration or the current of each seek control is made consistent.

[0211] By employing the above (A-1), it is possible to perform a seek operation that reaches the target data sector earlier, even when a certain actuator block is not accessing a data sector.

[0212] By adopting the above (A-2), it is possible to implement (A-1) when the HDD's power mode is idle, standby, or when the command queue for the HDD requested by the host is empty.

[0213] By adopting the above (A-3), it is possible to implement low-Jerk seek control and high-Jerk seek control that limit Jerk.

[0214] By adopting the above (A-4), it is possible to implement low-Jerk seek control and high-Jerk seek control with limited Jerk in a configuration with low overhead caused by processing time on the FW.

[0215] By employing the above (A-5), the command control side can rearrange commands corresponding to the seek control of high Jerk.

[0216] By employing the above (A-6), the command control side can rearrange commands corresponding to high-Jerk seek control. Without this, even with high-Jerk seek control, spin waits will occur without improving access performance.

[0217] By employing the above (A-7), the timing can be adjusted so that no data sector is written to or read from during the time given to the disturbance vibration caused by the high Jerk seek of a certain actuator block.

[0218] By employing the above (A-8), the timing can be adjusted so that no data sector is written to or read from during the time given to the disturbance vibration caused by the high Jerk seek of a certain actuator block.

[0219] By adopting the above (A-9), power consumption can be reduced while maintaining the same access performance.

[0220] By employing the above (A-10), it is possible to perform seek control adjustments that end the seek just before the moment the target data sector is about to be accessed.

[0221] By adopting the above (A-11), even if the current peak or acceleration peak is the same, the seek time can be changed by changing Jerk.

[0222] According to at least one embodiment or variation described above, a disk device with improved performance of multiple actuators can be provided.

[0223] Furthermore, the present invention is not limited to the above-described embodiments, and can be embodied by modifying the constituent elements during the implementation phase without departing from its spirit. Additionally, various inventions can be formed by appropriately combining the multiple constituent elements disclosed in the above embodiments. For example, several constituent elements may be deleted from all the constituent elements shown in the embodiments. Furthermore, constituent elements from different embodiments may be appropriately combined. Furthermore, the processes shown in flowcharts, timing diagrams, etc., can also be implemented by hardware including IC chips, digital signal processors (DSPs), software (programs, etc.) operating in computers including microcomputers, or a combination of hardware and software. Furthermore, the present invention is also applied when the claims are expressed as control logic, as a program containing instructions for computer execution, or as a recording medium readable by a computer containing said instructions. Furthermore, the names and terminology used are not limited; even other expressions are included in the present invention as long as they are substantially the same in content and spirit.

Claims

1. A disk device having two or more independently drivable actuator blocks, performing a low jerk seek control that limits jerk, which is jerk, characterized by, in the disk device, in a state where a first actuator block does not access a data sector of a disk, a second actuator block other than the first actuator block accesses the data sector of the disk by a seek control with high jerk.

2. The disk device according to claim 1, characterized by, in a case where a power mode of the first actuator block is not active or in a state where a command queue that holds a first command from a host to a disk access request to the first actuator block is empty, it is determined that the first actuator block is in a state where it does not access the data sector.

3. The disk device according to claim 1, characterized by, low jerk seek control parameters for performing the low jerk seek control that limits jerk and high jerk seek control parameters for performing the seek control with high jerk are held in a nonvolatile memory.

4. The disk device according to claim 3, characterized by, the low jerk seek control parameters and the high jerk seek control parameters held in the nonvolatile memory are expanded in a volatile memory at start-up.

5. The disk device according to claim 1, characterized by, in the low jerk seek control that limits jerk and the seek control with high jerk, each has a corresponding table of seek time with respect to seek distance for implementing command reordering.

6. The disk device according to claim 5, characterized by, the corresponding table is switched when switching from the low jerk seek control that limits jerk to the seek control with high jerk.

7. The disk device according to any one of claims 1 to 6, characterized by, in a period where a third actuator block is performing the high jerk seek control, when a fourth actuator block other than the third actuator block performs seek control, in a case where there is one or more fourth commands that can access a data sector at a time later than a seek end time of the third actuator block, a command to be next accessed is selected from the fourth commands.

8. The disk device according to claim 7, characterized by, in a period where a fifth actuator block is performing the high jerk seek control, when a sixth actuator block other than the fifth actuator block performs seek control, in a case where there is no command that can access a data sector by the fifth actuator block at a time later than a seek end time of the fifth actuator block, a command to be next accessed is selected from a command group, a one revolution time is inserted so that the access is performed at a time later than the seek end time of the fifth actuator block.

9. The disk device according to claim 8, wherein After the target data sector is determined in the seek control of the 7th actuator block, adjustment of the seek control of the 7th actuator block is performed so that the seek of the 7th actuator block ends before the time of access to the target data sector.

10. The disk device according to claim 9, wherein In the adjustment of the seek control of the 7th actuator block, any combination of time or acceleration level in the acceleration section, time or acceleration level in the deceleration section, and time in the constant velocity section of the actuator is used for the adjustment.

11. The disk device according to claim 10, wherein In the seek control, in the case of switching between the seek control with high jerk and the seek control with low jerk, the peak level of any of the acceleration and the current in the action characteristics of each seek control is made uniform.

Images