Disc device and head control method

Abstract

Provided are a magnetic disk device and a head control method that suppress vibration of a head by feedforward control. The magnetic disk device according to the present embodiment includes a feedforward control section that is input with a VCM control value and head position information that is information of a current position of a head, as input values, and outputs a MA control correction value for a micro-actuator, as an output value, the feedforward control section including a plant model processing section that includes a transfer function model that represents a relationship between the input values and the output value, the plant model processing section further including a plurality of sub-model processing sections, each of the sub-model processing sections generating a sub-model output value that is an output value for the input values that are input, based on a transfer function model that is respectively given, and the feedforward control section generating the output value based on a plurality of the sub-model output values.

Description

[0001] This application enjoys priority based on Japanese Patent Application No. 2022-151035 (filed on September 22, 2022). This application incorporates the entire contents of the basic application by reference. Technical Field

[0002] Embodiments of the present invention relate to a disk device employing feedforward control and a head control method. Background Technology

[0003] In disk drives such as hard disk drives (HDDs), during seek operations, the voice coil motor (VCM) accelerates / decelerates due to the seek current. Simultaneously, other resonant modes of the VCM are stimulated, resulting in residual vibrations that occur during settling. To suppress these residual vibrations by using feedforward control to reduce them from the seek current, the feedforward controller incorporates elements of a transfer function from the current input to the VCM to the head displacement.

[0004] However, one example of a mode that is prone to residual vibration in the VCM of an HDD is the coil torsion mode. This mode continues to vibrate due to its small damping coefficient, and thus manifests in the residual vibration after the actuator has moved significantly through seek control. However, when the transfer function from the current input to the VCM to the head position is measured, it is buried in the rigid body mode and the main resonance mode and is difficult to see clearly. Summary of the Invention

[0005] Embodiments of the present invention provide a disk device and a head control method for suppressing head vibration through feedforward control.

[0006] The disk device according to this embodiment includes: a head for reading data from and writing data to the disk; an arm for supporting the head; a voice coil motor (VCM) for controlling the arm to perform seek control on the head; a micro-actuator for finely adjusting the position of the head; a servo control unit for generating VCM control values ​​input to the VCM and MA control values ​​input to the micro-actuator; and a feedforward control unit that takes the input VCM control values ​​and information about the current position of the head (i.e., head position information) as input values ​​and outputs an MA control correction value for the micro-actuator as an output value. The feedforward control unit includes a controlled object (plant) model processing unit that includes a transfer function model representing the relationship between the input values ​​and the output values. The controlled object model processing unit further includes multiple sub-model processing units. Each sub-model processing unit generates a sub-model output value based on a given transfer function model. The sub-model output value is an output value for the input values. The feedforward control unit generates the output value based on the multiple sub-model output values. Attached Figure Description

[0007] Figure 1 This is a configuration diagram of the disk device according to the first embodiment.

[0008] Figure 2 This is an example of a block diagram of the servo control unit and related parts in the disk device according to the first embodiment.

[0009] Figure 3A This is an example of the transfer function of the head displacement relative to the input current to the VCM in the disk drive according to the first embodiment. Figure 3A (a) represents the gain characteristic. Figure 3A (b) represents the phase characteristic.

[0010] Figure 3B This is an example of the time response of each sub-model in the disk device according to the first embodiment to the seek current.

[0011] Figure 4 This diagram illustrates the processing steps of the servo control unit and related parts of the disk device according to the first embodiment.

[0012] Figure 5 This is an example of a block diagram of the servo control unit and related parts in the disk device according to the second embodiment.

[0013] Figure 6 This diagram illustrates the processing operation of the servo control unit and related parts in the disk device according to the second embodiment.

[0014] Figure 7 This is an example diagram of the actuator and related parts in the disk device according to the third embodiment.

[0015] Figure 8 This is an example of a block diagram of the servo control unit and related parts in the disk device according to the third embodiment.

[0016] Figure 9 This is an example of storing a database of sub-models applied to the feedforward control unit for each head in a disk device according to the third embodiment. Figure 9 (a) is an example where the "arm position" and "head mounting position" are set. Figure 9 (b) is an example where a "group" is defined. Figure 9 (c) is an example with three sub-model processing units.

[0017] Label Explanation

[0018] 1 Disk drive; 2 Host system; 10 HDA; 11 Disk; 12 SPM; 13 VCM; 14 Pivot; 15 Arm; 16 Suspension; 17 Microactuator; 18 Slider; 19 Head; 20 Driver IC; 30 Head Amplifier IC; 40 R / W Channel; 50 HDC; 60 Main Controller; 61 Servo Control Unit; 62 Read / Write Control Unit; 63 Feedforward Control Unit; 64 Parameter Update Unit; 70 Volatile Memory; 80 Buffer Memory; 90 Non-volatile Memory; 111 Power Supply; 112 FPC. Detailed Implementation

[0019] The embodiments will now be described with reference to the accompanying drawings. Furthermore, the drawings are illustrative and do not limit the scope of the invention.

[0020] (First Embodiment)

[0021] The disk device in this embodiment decomposes the controlled object model of the feedforward control of multiple actuators into multiple vibration modes such as rigid body mode, main resonance mode, and other vibration modes, defines the transfer function model for each vibration mode, and identifies / learns (trains) the parameters of the transfer function model.

[0022] Figure 1 This is a configuration diagram of the disk device involved in the implementation method.

[0023] Disk device 1 is a storage device equipped with a disk (hereinafter sometimes simply referred to as a disk) for reading and writing data, and includes a microprocessor or other computer components. Disk device 1 outputs data to host system 2 or writes data input from host system 2 to the disk based on commands received from the connected host system 2.

[0024] The disk device 1 in this embodiment includes a feedforward controller for controlling the microactuator 17. The feedforward controller internally has a controlled object model that ranges from the control value of the VCM as an input value to the head displacement (also known as the control correction value of the head position) as an output value. The controlled object model of the feedforward controller internally connects multiple sub-models in parallel, and uses the output of at least one of the multiple sub-models for feedforward control of the microactuator 17.

[0025] The host system 2, such as a personal computer, outputs data read / write commands to the disk device 1. Additionally, the host system 2 may have a program for testing the disk device 1, and may output test signals to test the disk device 1.

[0026] HDA10 is referred to as a head assembly, which houses a disk 11, a spindle motor (hereinafter sometimes also called SPM) 12, an arm 15 with a head 19 mounted on it, and a voice coil motor (hereinafter sometimes also called VCM) 13, etc., within a housing. Although this embodiment shows an example in which one disk 11 and one head 19 are provided, more than one may be provided.

[0027] HDA10 is fixed to the disk drive 1 by housing support portions 101A, 101B, 101C, and 101D (hereinafter referred to as housing support portion 101 unless otherwise specified). In addition to fixing HDA10, housing support portion 101 also has the function of mitigating the impact of vibration of the disk drive 1 as a whole on HDA10 by having spring characteristics.

[0028] Disk 11 is a disk-type magnetic rotating disk storage medium. Within the data area where data can be written (sometimes referred to as write), there is a user data area accessible to the user and a system area for writing information required for system management. Hereinafter, the direction orthogonal to the radial direction of disk 11 will be referred to as the circumferential direction. Disk 11 is mounted on a spindle motor 12 and rotates under the drive of the spindle motor 12.

[0029] Disk 11 has multiple tracks. Figure 1 In the example shown, three tracks TR1, TR2, and TR3 (hereinafter referred to as tracks TR unless otherwise specified) are depicted. However, multiple tracks are arranged concentrically around the spindle motor 12 within the data area. During data reading and writing (read / write) on the execution disk 11, seek control, tracking control, etc., move the head 19 to the track TR containing the data to be read / written (sometimes also called target data), and the head 19 performs the read / write operation. Sometimes, the track TR containing the target data is also referred to as the target track.

[0030] Additionally, servo information is written to disk 11 for purposes such as position detection of head 19. The servo information is set at a predetermined position (called the servo area) in the circumferential direction of disk 11. The servo information is general in nature and detailed explanations are omitted. Figure 1 In the example, three servo regions SVA1, SVA2, and SVA3 (hereinafter referred to as servo regions SVA unless otherwise specified) are shown as examples of servo regions. However, generally, servo regions SVA are set at equal intervals along the entire circumference of disk 11, and servo information is written to the servo regions SVA of each track TR. The disk device 1 can detect the current position of the head 19 based on the servo information read from the head 19.

[0031] The spindle motor 12 is a support for the disk 11 and is mounted on the disk housing, etc. The disk 11 rotates by the spindle motor 12.

[0032] VCM13 is a voice coil motor type actuator used for moving arms 15, etc. VCM13 controls the movement of arms 15, etc., based on input current or voltage. VCM13 includes a VCM coil 131 for power transmission.

[0033] Pivot 14 is a bearing used to support the arm 15 and so on, and to enable it to rotate.

[0034] Arm 15 is the arm that supports slider 18 and head 19, and transmits power from VCM13 to head 19, causing head 19 to move to the target track TR.

[0035] The suspension 16 controls the movement of the connected microactuator 17.

[0036] Microactuator 17 (sometimes also called MA17) performs high-precision position adjustments, such as tracking control of head 19, based on input current or voltage. While the microactuator 17 has a general function and is omitted in detail, it performs fine-tuning of the position of head 19 after seek control and / or tracking control of the target track after seek control. Stabilization refers to the time after head 19 has been moved via seek control until the positioning error of the target track, including the effects of vibration caused by seek control, falls below a certain threshold. Once the vibration of head 19 has become sufficiently small through stabilization, data read / write control and / or tracking control are performed.

[0037] The slider 18 is the part that supports the head 19 and transmits power.

[0038] Head 19 is the part that writes data to disk 11 and reads data recorded on the data tracks of disk 11. In cases where special distinction is required, the head that writes data to disk 11 is called the write head 19W, and the head that reads data recorded on the data tracks of disk 11 is called the read head 19R.

[0039] Power supply 111 is the power supply that supplies power to coil 131 of VCM13.

[0040] FPC112 is a flexible printed circuit, a cable that supplies current or voltage, connected near the root of power supply 111 and VCM13. FPC112 vibrates according to the operation of VCM13, and may therefore be a cause of residual vibration. FPC112 sometimes retains its inherent vibration mode (called FPC mode), which will be explained later.

[0041] The driver IC20 outputs current or voltage for driving and controlling the SPM12, VCM13, micro actuator 17, etc., according to the control from the HDC50, servo control unit 61, etc.

[0042] The head amplifier IC30 includes a read amplifier and a write driver. The read amplifier amplifies the read signal read from disk 11 and outputs it to R / W channel 40. The write driver outputs a write current corresponding to the signal output from R / W channel 40 to head 19.

[0043] The R / W channel 40 controls the head amplifier IC 30 to read data from the disk 11 and write data to the disk 11 according to instructions from the HDC 50, the main controller 60, etc. The R / W channel 40 receives read data signals from the head amplifier IC 30 and extracts read data, or generates write data signals based on commanded write data and outputs them to the head amplifier IC 30. Additionally, the R / W channel 40 has the function of measuring the signal quality of the read data received from the head amplifier IC 30. The R / W channel 40 can also extract the position information of the head 19 based on servo information signals received from the head amplifier IC 30.

[0044] HDC50 is the interface between disk drive 1 and host system 2, and it is the hard disk controller that controls various parts of disk drive 1. HDC50 is composed of a computer such as a CPU, or it can be composed of an IC chip with computer functionality. HDC50 receives commands from host system 2, such as write commands to write data to disk 11 and read commands to read data from disk 11. Based on the received commands, HDC50 controls various parts of disk drive 1, or performs data transfer between host system 2 and R / W channel 40. HDC50 can also control the reading and writing of data to volatile memory 70, buffer memory 80, and non-volatile memory 90.

[0045] The main controller 60 is the main controller that controls the various parts of the disk drive 1. It may be composed of a computer, such as a microprocessor, or an IC chip with computer functions. The various processes of the main controller 60 may be executed by software programs (including firmware), or they may be provided as hardware or a combination of software and hardware.

[0046] The servo control unit 61 controls the head 19 of the data, for example, based on commands received from the host system 2. For example, in order to move the head 19 to the target position (target track), the servo control unit 61 performs seek control via the driver IC 20 to control the VCM 13. In addition, the servo control unit 61 controls the microactuator 17 via the driver IC 20 to perform tracking control of the head 19 after seek control.

[0047] More specifically, the servo control unit 61 determines the input current value or input voltage value as the control value for seek control and / or tracking control of the head 19 based on the target position of the head 19 determined by the command received from the host system 2 and the current position of the head 19 received from the R / W channel 40, and outputs it to the VCM 13 and the micro actuator 17.

[0048] The read / write control unit 62 selects the destination for writing data (e.g., information about the data sector, track, etc. of disk 11) and controls the writing operation to disk 11. Additionally, the read / write control unit 62 controls the reading operation from disk 11, or controls the processing of the read data read from disk 11 and outputs it to the HDC50, etc.

[0049] The feedforward control unit 63 includes a controlled object model comprising multiple sub-models and various computing units, and performs feedforward control on the microactuator 17. The sub-models are, for example, digital filters defined by transfer function models based on vibration modes, etc. Based on the current position of the head 19 and the current value used as the control value for the VCM 13, the feedforward control unit 63 generates control correction values ​​for the microactuator 17 through various data processing steps and outputs them to the servo control unit 61.

[0050] In the servo control unit 61, the input control value for the microactuator 17 is added to the control correction value input from the feedforward control unit 63, and the result is output to the microactuator 17 as the final control value for the microactuator 17.

[0051] Furthermore, although this embodiment shows a configuration where the servo control unit 61 and the feedforward control unit 63 are separate, they can also be functionally combined into one unit, the servo control unit 61. Additionally, in this embodiment, the feedforward control unit 63 receives control values ​​for the VCM13 from the servo control unit 61 and receives the current position of the head 19 from the R / W channel 40.

[0052] The parameter update unit 64 performs parameter identification and parameter updating of the sub-model based on the control value u_v of the control VCM13, the current position of the head 19, or values ​​calculated from these values. Parameter identification means obtaining parameters, and parameter updating means updating the (identified) parameters obtained at one time, although there are cases where both are used in the same sense. The parameter update unit 64 performs parameter identification or updating using any parameter fitting method, such as the least squares method. Various general techniques can be used for parameter fitting, and detailed descriptions are omitted. Alternatively, the parameter update unit 64 may use successive least squares methods, etc., to successively perform parameter identification or updating each time the control value u_v of the control VCM13 and / or the current position of the head 19 is obtained.

[0053] The parameter update unit 64 can also store the identified or updated parameters in volatile memory 70, non-volatile memory 90, etc., and retrieve and use them when the next parameter is identified or updated. In addition, when updating parameters sequentially, the parameter update unit 64 can store the updated parameters in buffer memory 80, etc., and the feedforward control unit 63 can also retrieve the updated parameters stored in buffer memory 80 and apply the retrieved parameters to the sub-model to perform data processing.

[0054] Alternatively, the VCM control value u_v and the current position of header 19 can be obtained in advance as measured values, or simulated data can be obtained as test data. When the test data is input, the feedforward control unit 63 is activated during the calibration of the disk device 1, causing the parameter update unit 64 to identify and update the parameters. Furthermore, the parameter update unit 64 can also perform parameter identification or parameter updates during the use of the disk device 1 (in operation) using the obtained VCM control value u_v and the current position of header 19.

[0055] Volatile memory 70 is a semiconductor memory that loses its stored data when the power supply is disconnected. Volatile memory 70 stores data required for processing in disk drive 1. Volatile memory 70 is, for example, DRAM (Dynamic Random Access Memory) or SDRAM (Synchronous Dynamic Random Access Memory).

[0056] The buffer memory 80 is a semiconductor memory that temporarily records data transmitted and received between the disk drive 1 and the host system 2. Furthermore, the buffer memory 80 may be integrated with the volatile memory 70. The 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).

[0057] Non-volatile memory 90 is a semiconductor memory that records stored data even when the power supply is interrupted. Non-volatile memory 90 is, for example, a NOR or NAND flash read-only memory (FROM).

[0058] The various functions of R / W channel 40, HDC50, main controller 60, etc. can also be used as system controllers, for example, implemented on a large-scale integrated circuit (LSI) called System-on-a-Chip (SoC), which integrates multiple components on a single chip.

[0059] Figure 2 This is an example of a block diagram of the servo control unit and related parts in the disk device according to the first embodiment.

[0060] Controller 610 represents a block diagram of a controller that includes the functions of a servo control unit 61 and a feedforward control unit 63, and represents a transfer function or a controller having the transfer function that takes the target position r of the head 19 as input and the current head position y of the head 19 as output.

[0061] Cv611 is a controller that controls VCM13, and is included, for example, in the servo control unit 61. Based on inputs such as the difference between the target position r and the current position y of head 19 (called head position error) e, Cv611 generates a control value u_v for controlling VCM13 and outputs it to VCM13. Cv611 represents a transfer function that takes the head position error e as input and the control value u_v as output, or a controller having that transfer function. Alternatively, the displacement error, which replaces the head position y and uses the target displacement amount of the head position relative to the target position r of head 19, can be set as e.

[0062] Cm612 is a controller that controls the microactuator 17, and is included, for example, in the servo control unit 61. Cm612 generates a control value u_m1 based on inputs such as the head position error e, and outputs it to the microactuator 17. Cm612 represents a transfer function that takes the head position error e as input and the control value u_m1 as output, or a controller having that transfer function. Alternatively, the displacement error, which replaces the head position and uses the target displacement amount of the head position relative to the target position r of the head 19, can be set as e.

[0063] Pv131 and Pm171 represent the transfer functions of VCM13 and microactuator 17, respectively.

[0064] Controller 613 is a block diagram of a controller equivalent to feedforward control unit 63, consisting of a controlled object model containing multiple sub-models. Each sub-model is a digital filter that processes digital sample values; the relationship between input and output (transfer function) is modeled using parameters, formulas, etc. Hereinafter, the modeled transfer function will sometimes be referred to as the transfer function model.

[0065] The controller 613 in this embodiment includes three sub-models F1, F2, and F3, which take the control value u_v and / or the head position y as inputs and output a control value u_m2 for the microactuator 17. The control value u_m2 can also be, for example, a current value or a voltage value. Furthermore, when the control value u_v is input, the sub-models F1, F2, and F3 generate output values ​​(called sub-model output values) based on their respective transfer function models. In addition, sometimes the transfer function model itself is also referred to as a sub-model.

[0066] In controller 613, an error value is generated as the difference between the output value of each sub-model and the head position y, etc.

[0067] The first sub-model processing unit 6131 generates an error value e1 based on the sub-model output value y1 and the head position y generated by the transfer function model of sub-model F1 with the control value u_v as input.

[0068] The second sub-model processing unit 6132 generates an error value e2 based on the sub-model output value y2 and the error value e1 generated by the transfer function model of sub-model F2 with the control value u_v as input.

[0069] The third sub-model processing unit 6133 generates an error value e3 based on the sub-model output value y3 and the error value e2 generated by the transfer function model of sub-model F3, with the control value u_v as input. Furthermore, the third sub-model processing unit 6133 generates and outputs a control value u_m2 for the micro-actuator 17 based on the sub-model output value y3.

[0070] In the controller 613, although the case with three sub-models is taken as an example, if there are two sub-models, at least two sub-model processing units are required: a first sub-model processing unit 6131 and a third sub-model processing unit 6133. In this case, the third sub-model processing unit 6133 uses the error value e1, which is the output of the first sub-model processing unit 6131, instead of the error value e2.

[0071] During the seek control of the head 19, various vibrations will occur and affect the position of the head 19. For example, consider any vibration mode such as rigid body mode, main resonance mode, coil resonance mode, FPC mode, shell support system mode, etc. The sub-model of this embodiment is determined based on these vibration modes.

[0072] Figure 3A This is an example illustrating the frequency response of the head displacement in the disk drive according to the first embodiment relative to the transfer function of the input current to the VCM. Figure 3A (a) represents the gain characteristic. Figure 3A (b) represents the phase characteristic.

[0073] The transfer function TF0 represents an example of the transfer function for the head displacement vy under various vibration modes for a head position y given an input value u_v to VCM13. The head displacement vy is assumed to be the head displacement caused by residual vibration. Transfer functions TF1, TF2, and TF3 are the transfer functions for each vibration mode decomposed from the transfer function TF0.

[0074] The transfer function TF1 represents the transfer function of the rigid body mode. The rigid body mode is the vibration generated by the movement of head 19 through VCM13, and it is a mode that will definitely be generated during seek control.

[0075] The transfer function TF2 represents the transfer function of the main resonance mode. The main resonance mode is a vibration mode generated primarily by mechanisms including arm 15. The main resonance mode is the main cause of the position error of head 19, and is a vibration mode that is preferably suppressed by stabilizing the phase using VCM13, rather than by suppressing it using microactuator 17.

[0076] The transfer function TF3 represents the transfer function of the vibration modes (referred to as other vibration modes) obtained by removing TF1 and TF2 from the transfer function TF0. Other vibration modes may include coil resonance mode, FPC mode, shell support system mode, etc.

[0077] Figure 3B This is an example of the time response of each sub-model in the disk device according to the first embodiment to the seek current.

[0078] Characteristic TC2 is an example of the time response of the second sub-model processing unit 6132 (transfer function TF2), and characteristic TC3 is an example of the time response of the third sub-model processing unit 6133 (transfer function TF3).

[0079] The coil resonance mode is a vibration mode generated by the VCM coil 131 as the main source, and it is a vibration mode that should be suppressed by the microactuator 17.

[0080] The FPC mode is a vibration mode mainly generated by the FPC112, which varies with the position of the VCM13 powered by the FPC112, and is a vibration mode that should be suppressed by the microactuator 17.

[0081] The shell support system mode is a vibration mode generated primarily by the shell support part 101. Its impact is usually reduced through functions such as rotational vibration feed-forward control (RVFF).

[0082] The controlled object model of the feedforward control unit 63 in this embodiment includes multiple sub-models that take into account the vibration modes described above. The parameter identification method for the sub-models in this embodiment can be any method, as illustrated below.

[0083] (Parameter identification method 1)

[0084] This represents an example of sub-modeling all considered vibration modes as a transfer function model for the controlled object model of the feedforward control unit 63, and identifying the parameters of the sub-model.

[0085] For example, suppose there are M vibration modes under consideration. The sub-models for the M vibration modes are included as a whole model (sometimes called the global model). The sub-models are taken as examples in one case. The input value of the global model at time t is represented as u(t), and the output value is represented as y(t). The transfer functions F(s) of the respective Laplace transforms U(s) and Y(s) are shown below.

[0086]

[0087] Where s represents the Laplace operator, ζ i ω i k i This represents the attenuation rate, angular frequency, and residual number of each mode. Hereinafter, the vibration modes of the overall model will be referred to as the overall vibration modes.

[0088] The physical model of the transfer function is represented using the Laplace operator in Equation 1, but in the feedforward control unit 63, for example, in actual processing, digital sample values ​​of the discrete-time system obtained by sampling y(t) obtained from the above equation at sampling intervals Ts are used. In particular, the transfer function model obtained by performing an sz transform on the transfer function model F(s) is represented as F(z). For example, if j is set to an integer, the discrete-time input and output at time t = j × Ts are represented as u(j) and y(j), respectively, and their respective z-transforms are U(z) and Y(z). The output Y(z) when the sub-model with the transfer function model F(z) is input with U(z) is shown as Y(z) = F(z)U(z). By using the difference equation obtained from the transfer function model F(z) by the inverse z-transform, y(j) can be calculated successively from u(j).

[0089] The parameter update unit 64 performs parameter identification on the transfer function model F(s) described in Equation 1. More specifically, the parameter update unit 64 obtains the measured value of the frequency response characteristic of the head position y for the output u_v (VCM input) of Cv611 as the measured value for F(z). Based on the obtained measured value, the parameter update unit 64 performs parameter identification using various parameter fitting methods. For example, when using the least squares method, the parameter update unit 64 updates the parameters to minimize the square of the difference between y(j) and the head position y.

[0090] The parameter update unit 64 can also acquire measured values ​​by measuring the actual frequency response characteristics. Alternatively, measured values ​​obtained from an external source can be stored in a non-volatile memory 90, and the parameter update unit 64 retrieves the measured values ​​from the non-volatile memory 90 during parameter updates. The transfer function model of the identified overall model can also be further modeled by separating specific resonance modes as needed.

[0091] (Parameter identification method 2)

[0092] This indicates the application of the controlled object model to the feedforward control unit 63. Figure 2 The example shown is based on the model of the feedforward controller 613, illustrating the parameter identification of sub-models. In this example, after pre-defining the transfer function models of sub-models F1, F2, and F3, parameter identification is performed for each sub-model. In this example, the overall vibration mode is decomposed into three modes, such as rigid body mode, main resonance mode, and other specific vibration modes, and their respective transfer function models are defined. Furthermore, there are various methods for mode decomposition; this example is not limited to any particular method. The following describes the sub-models in this example based on the mode form using the Laplace operator.

[0093] The vibration mode of sub-model F1 is set as rigid body mode. The following is the transfer function F1(s) for input value U(s) and output value Y1(s).

[0094]

[0095] The vibration mode of sub-model F2 is set as the main resonance mode. The following is the transfer function F2(s) for the input value U(s) and the output value Y2(s).

[0096]

[0097] Set the vibration mode of sub-model F3 to other modes (assuming there are M-2 modes). The following is the transfer function F3(s) for the input value U(s) and the output value Y3(s).

[0098]

[0099] The following describes the parameter identification of the model for controller 613. Furthermore, similar to parameter identification method 1, the transfer function model F(z) obtained by the sz transformation of F(s) is used in the calculations of the feedforward control unit 63.

[0100] In the controller 613 of this embodiment, the vibration modes of F1 and F2 are set to rigid body mode and main resonance mode, respectively, and the adaptive parameters of their respective transfer function models F1(z) and F2(z) are all fixed by setting parameters determined in advance based on measured values, etc. In the controller 613, the vibration mode of F3 is set to other modes, and the transfer function model is set to F3(z). The update timing of the target head position r and head position y can be, for example, the acquisition interval of the servo information obtained by the head 19 (i.e., the sampling interval of the servo information), or any interval such as the time interval determined according to the multi-rate, etc.

[0101] Figure 4 This diagram illustrates the processing operation of the servo control unit and related components of the disk drive according to the first embodiment. The diagram is for illustrative purposes. Figure 2 The block diagram shown depicts data flow and processing, not a flowchart representing a fixed processing order. Therefore, it is not always necessary to follow the order. Figure 4 The processes are executed in the order shown, and each process can be performed sequentially or simultaneously.

[0102] The servo control unit 61 calculates the head position error e based on the target position r of the head 19 determined by commands received from the host system 2 and the current head position y of the head 19 received from the R / W channel 40, and inputs it to Cv611 and Cm612 (step S101). Cv611 generates and outputs the control value u_v for controlling VCM13 based on the input head position error e (step S102). In addition, Cm612 generates and outputs the control value u_m1 for controlling the microactuator 17 based on the input head position error e (step S1021).

[0103] The feedforward control unit 63 (equivalent to controller 613) generates and outputs the control correction value u_m2 from the output value y3 of the input value u_v for the sub-model F3, after passing through the filter (-1 / Pm) (step S103). Here, Pm is set to the transfer function of Pm171.

[0104] The control value u_m (=u_m1+u_m2) obtained by adding the control correction value u_m2 to the control value u_m1 generated by Cm612 is input to the micro actuator 17 (equivalent to Pm171) (step S104).

[0105] On the other hand, in the feedforward control unit 63, the first sub-model processing unit 6131 receives the current head position y of the head 19 from the R / W channel 40 and calculates the error value e1 (=r-y1) using the output value y1 of the sub-model F1 (step S108). The second sub-model processing unit 6132 calculates the error value e2 (=e1-y2) based on the output value y2 and the error value e1 (step S109). The third sub-model processing unit 6133 calculates the error value e3 (=e2-y3) based on the output value y3 and the error value e2 (step S110). The parameter update unit 64 identifies the parameters of the transfer function model of the sub-model F3 based on the error value e3 (step S111).

[0106] In step S111, when the parameter update unit 64 receives the error value e3 from the feedforward control unit 63, it identifies the parameters of the transfer function model of the sub-model F3 using any parameter fitting method such as the least squares method. For example, when using the least squares method, the parameter update unit 64 determines k as the adaptive parameter of the sub-model F3 in a way that minimizes the square of the error value e3. i ζ i ω i .

[0107] After parameter update unit 64, the parameters of the transfer function model of sub-model F3 can be updated sequentially using the same process as described above, for example, by the interval at which servo information is acquired (i.e., the sampling interval of servo information).

[0108] Through the above process, the parameters of the sub-models of the feedforward controller 613 can be identified. A feature of this embodiment is that, in the parameter update of the updated sub-model (equivalent to F3), error values ​​(equivalent to error values ​​e1 and e2) calculated based on the outputs of other sub-models (equivalent to F1 and F2) are used as input. That is, the feedforward controller 613 of this embodiment is characterized by using, as... Figure 4 As shown in steps S108 to S111, an error value e1 is generated based on the output y1 of sub-model F1 and the current position y of the head, an error value e2 is generated based on the output y2 of sub-model F2 and the error value e1, and an error value e3 is generated based on the output y3 of sub-model F3 and the error value e2, thereby cascading the output of each sub-model to reflect in other sub-models.

[0109] (Parameter identification method 3)

[0110] The transfer function model F3(s) of sub-model F3 can also be a transfer function of any degree N, instead of the pattern form shown in parameter identification method 2 above.

[0111]

[0112] Where s is the Laplace operator, and let a N a N-1 a N-2 ..., a1, a0, b N b N-1 b N-2 b1, b0, ..., b0 are adaptive parameters. Regarding the parameter identification method, the same method as shown in parameter identification method 2, i.e., the least squares method or any parameter fitting method, is used to identify the parameters of the transfer function model of sub-model F3. For example, when using the least squares method, the parameter update unit 64 determines a as the adaptive parameter of sub-model F3 in a way that minimizes the square of the error value e3 generated in the same way as in parameter identification method 2. N a N-1 a N-2 ..., a1, a0, b N b N-1 b N-2 b1, b0.

[0113] (Parameter identification method 4)

[0114] Alternatively, the transfer function model of sub-model F3 can be modeled using a general ARX model (Auto-Regressive with eXogenous model) such as the following (Equation 6).

[0115]

[0116] Here, let's set it as

[0117] A(q -1 )=1+a1q -1 +a2q -2 +……a n q -n (Equation 6-1)

[0118] B(q -1 )=1+b1q -1 +b2q -2 +……b n q -n (Equation 6-2)

[0119] Additionally, q represents the sampling delay of the discrete-time system, and j represents the time t = j × Ts (or the sampling order) with respect to the sampling interval Ts. y3(j), u_v(j), and e3(j) represent the output y3 to sub-model F3, the input u_v, and the error value e3 given to the third sub-model processing unit 6133 at the time t when y3(j) is calculated, respectively.

[0120] In this example, let a1, a2, ..., a1 in (Equation 6-1) and (Equation 6-2) be... n b1, b2, ..., b n For example, in the case of successive use of the least squares method, the parameter update unit 64 determines and updates the adaptive parameters a1, a2, ..., a3 as sub-model F3 at each time t = j × Ts in a manner that minimizes the square of the error value e3(j) generated in the same way as parameter identification method 2. n b1, b2, ..., b n The parameter update timing can be any timing, such as the timing for obtaining servo information.

[0121] The above parameter identification or updates can be performed using test signals during design, manufacturing, and adjustment, or, for example, readjusted on-site after delivery. As test signals, arbitrary data such as the VCM control value u_v (which serves as the input value to the feedforward control unit 63, equivalent to controller 613) and analog or measured data of the head position y can also be used. If the disk drive 1 has the function of generating test signals, calibration and adjustment can be performed using test signals, for example, when the head 19 is not accessing the disk 11.

[0122] Alternatively, the input test signal and parameter identification method can be arbitrarily changed according to each sub-model. Regarding the input signal, for example, it can be based on a trapezoidal wave, and the height, width, slope, etc. of the trapezoidal wave can be changed to make it easier to identify the vibration mode as the target.

[0123] (Parameter identification method 5)

[0124] Alternatively, the parameters identified in parameter identification method 4 above can be used as feedforward control, and the parameters of the sub-model that generates the MA control correction value output to MA17 can be updated adaptively.

[0125] For example Figure 2 The parameters of the transfer function models of sub-models F1 and F2 are set to fixed, and the parameters of the transfer function model are only adaptively updated for sub-model F3. In this method, the parameter update of the adaptively updated sub-model (equivalent to F3) is also related to... Figure 4 Similarly, steps S108 to S111 reflect the outputs of other sub-models (equivalent to F1 and F2) in a cascading manner. The update timing during adaptation can be, for example, the timing of acquiring servo information, which can be implemented using test signals during design, manufacturing, and adjustment, or in applications such as seek control using the measured values ​​of VCM control value u_v and head position y.

[0126] Furthermore, the updated parameters are stored, for example, in non-volatile memory 90. When the power supply to the disk drive 1 is connected, the feedforward control unit 63 and the parameter update unit 64 can, for example, obtain the latest parameters from the non-volatile memory 90. The feedforward control unit 63 can perform sub-model-based data processing based on the obtained parameters, and the parameter update unit 64 can perform further parameter updates based on the obtained parameters.

[0127] According to this embodiment, by identifying and updating the parameters of the sub-model of the feedforward control unit 63, feedforward control that takes into account resonance modes and the like manifested in residual vibrations is achieved.

[0128] (Second Implementation)

[0129] In this embodiment, an example is shown where, as in parameter identification method 2, all parameters of the sub-models of the feedforward control unit 63 are adaptively updated after each model is predefined, and the outputs of other sub-models are used in the updates of sub-models F2 and F3. The configuration of the disk device in this embodiment is similar to... Figure 1 The structure is the same, but to distinguish it from other implementations, the labels added to the components are different. Figure 1 The label will be followed by an "A". For functions that do not differ significantly, the description will be omitted.

[0130] Figure 5 This is an example of a block diagram of the servo control unit and related parts in the disk device according to the second embodiment. Figure 5 Composition and Figure 2 The structure is the same, but to distinguish it from other implementations, in Figure 2 The designation is followed by an "A". Below, descriptions are omitted for functions without significant differences; only descriptions of functions that are similar to or different from the designation are provided. Figure 2 The difference in the controller 613A will be explained.

[0131] The controller 613A contains three sub-models F1, F2, and F3, but adaptively updates all sub-models and uses the outputs of other sub-models in the updates of sub-models F2 and F3.

[0132] Similar to the first embodiment, sub-model F1 is set as a rigid body mode, sub-model F2 is set as a main resonance mode, and the transfer function models of their respective models are set as those described in (Equation 2) and (Equation 3) above. Sub-model F3 can also be used as other modes, using any one of (Equation 4), (Equation 5), and (Equation 6).

[0133] Figure 6 This diagram illustrates the processing operation of the servo control unit and related components in the disk device according to the second embodiment. The diagram is for illustrative purposes. Figure 5The block diagram shown depicts data flow and processing, not a flowchart representing a fixed processing order. Therefore, it is not always necessary to follow the order. Figure 6 The processes are executed in the order shown, and each process can be performed sequentially or simultaneously.

[0134] The servo control unit 61A calculates the head position error e based on the target position r of the head 19A determined by commands received from the host system 2 and the current head position y of the head 19A received from the R / W channel 40A, and inputs it to Cv611A and Cm612A (step S201). Based on the input head position error e, Cv611A generates and outputs the control value u_v for controlling VCM13A (step S202). Furthermore, based on the input head position error e, Cm612A generates and outputs the control value u_m1 for controlling the microactuator 17A (step S2021).

[0135] The feedforward control unit 63A (equivalent to controller 613A) receives the current head position y of head 19A from R / W channel 40A. The first sub-model processing unit 6131A calculates the error value e1 (=r-y1) using the output value y1 of sub-model F1 for the input value of control value u_v (step S211). When the parameter update unit 64A receives the error value e1 from the feedforward control unit 63A, it identifies the parameters of the transfer function model of sub-model F1 using an arbitrary parameter fitting method (step S212). In step S212, for example, when using the least squares method, the parameter update unit 64A determines the adaptive parameters of sub-model F1 in a way that minimizes the square of the error value e1.

[0136] The second sub-model processing unit 6132A calculates the error value e2 (=e1-y2) based on the output value y2 and the error value e1 of the sub-model F2 (step S213). When the parameter update unit 64A receives the error value e2 from the feedforward control unit 63A, it identifies the parameters of the transfer function model of the sub-model F2 using an arbitrary parameter fitting method (step S214). In step S214, for example, when using the least squares method, the parameter update unit 64A determines the adaptive parameters of the sub-model F2 in a way that minimizes the square of the error value e2.

[0137] The third sub-model processing unit 6133A calculates the error value e3 (=e2-y3) based on the output value y3 and the error value e1 of the sub-model F3 (step S215). When the parameter update unit 64A receives the error value e3 from the feedforward control unit 63A, it identifies the parameters of the transfer function model of the sub-model F3 using an arbitrary parameter fitting method (step S216). In step S216, for example, when using the least squares method, the parameter update unit 64A determines the adaptive parameters of the sub-model F3 in a way that minimizes the square of the error value e3.

[0138] After parameter update unit 64A, the adaptive parameters of the transfer function models of sub-models F1, F2, and F3 can be updated sequentially using the same process, for example, with the interval of acquiring servo information (i.e., the sampling interval of servo information).

[0139] Through the above process, parameters can be identified and updated for the three sub-models of controller 613A.

[0140] (Third Implementation)

[0141] This embodiment illustrates an example where a disk device comprises multiple disks, multiple heads for reading and writing data to the multiple disks, and multiple arms that drive them, allowing selection of a sub-model for each head. The disk device configuration in this embodiment, apart from the HDA10, is similar to... Figure 1 The structure is the same, but to distinguish it from other implementations, the labels added to the components are different. Figure 1 The label will be appended with "B". For functions that do not differ significantly, the description will be omitted.

[0142] Figure 7 This is an example diagram of the actuator and related parts in the disk device according to the third embodiment.

[0143] VCM13B can control 5 arms 15B-N h (Unless otherwise specified, it is referred to as arm 15B) and mounted on pivot 14B. Arm 15B-N h Heads 19B-N, which are the objects to be controlled, are respectively installed on them. h (Unless otherwise distinguished, it is referred to as Head 19B).

[0144] For example, arm 15B-1 is an arm with head 19B-1 mounted on it, used to control and move head 19B-1. Alternatively, there are cases like arm 15B-2, which has two heads mounted vertically, namely heads 19B-21 and 19B-22. Arm 15B-2 controls and moves both heads separately. Similarly, arm 15B-3 is an example of two heads mounted vertically, namely heads 19B-31 and 19B-32, and arm 15B-4 is an example of two heads mounted vertically, namely heads 19B-41 and 19B-42. Arm 15B-5 is an example of one head 19B-5 mounted on it. That is, in this example, N h =8.

[0145] In addition, although Figure 7 The text does not include labels, but it does include... Figure 1 Other functions shown, such as suspension 16, microactuator 17, slider 18, etc., are also set up in the same way as each head 19B. Referring to them, the additional head identification number N is added to the label. hTo distinguish them from other implementations. For example, microactuators 17B-31, 17B-32 represent multiple actuators of control heads 19B-31, 19B-32.

[0146] In this embodiment, considering that the effect of residual vibration on head 19B varies depending on the installation position of arm 15B and / or head 19B, a model of the transfer function from the current control value input to VCM13B to the head position is defined for each head 19B, and parameters are identified or updated respectively.

[0147] Figure 8 This is an example of a block diagram of the servo control unit and related parts in the disk device according to the third embodiment.

[0148] In this embodiment, the configuration of each of the feedforward controllers 613B, which are configured with a number of heads 19B, is different from that of the feedforward controllers 613B. Figure 2 and Figure 5 It is the same, but to distinguish it from other implementations, the labels added to the components are... Figure 2 The label is appended with "B". For functions without significant differences, the following explanations are omitted. Additionally, feedforward controllers 613B are configured for each of the eight heads 19B, and the head identification number N is appended to the label. h To distinguish them.

[0149] Each feedforward controller 613B-N h The inputs u_v are respectively input to the VCM control input and are controlled by each feedforward controller 613B-N. h Output u_m2_N h u_m2_N h Equivalent to Figure 2 u_m2 represents the value according to each feedforward controller 613B-N h The output u_m2 (feedforward input). For example, the feedforward controller 613B-31 takes the input VCM control input u_v as input and outputs u_m2_31.

[0150] In the disk drive 1B of this embodiment, controllers 610B are provided for each of the eight heads 19B, and a header identification number N is added to the label. h To differentiate them. For example, the control head 19B-N h Controller 610B-N h Specifically, controllers 610B-21 and 610B-22 are respectively set for heads 19B-21 and 19B-22. Without specifically distinguishing controller 610B-N... h In this case, it is represented as controller 610B. Additionally, for controller 610B-N... h The internal block is set to be parallel. Figure 2 Similarly, the controller 610, in the case of reference, is set to be for... Figure 2 The labels shown are appended with BN. h The label. For example Figure 2 Controller 613 within controller 610, in controller 610B-N h In, for example, controller 613B-N h This is how it is expressed.

[0151] Each controller 610B and Figure 2 Controller 610 Figure 5 The controller 610A is the same, representing the inputs to the head 19B-N of the controlled object, respectively. h Target position r0N h The transfer function of the head position y_s and the output control inputs u_m1+u_m2 and u_v, or a controller having such a transfer function. For example, for the controller 610B-31 with head 19B-31, the target position r031 is input, and the output control inputs u_m1_31+u_m2_31 and u_v31 are output.

[0152] Figure 9 This is an example of storing a database of sub-models applied to the feedforward control units for each head in a disk device according to the third embodiment. The database is stored, for example, in non-volatile memory 90.

[0153] exist Figure 9 In the database example of (a), row-wise storage is used with header 19B-N h Associated information. "Head Identification Number N" h "Indicates the correct head 19B-N" h The assigned identification head information (head identification information), "arm position" indicates head 19B-N h The position of the connected arm 15B, "head mounting position" indicates head 19B-N h The vertical position, "adaptive model" indicates the relationship with controller 610B-N h The feedforward control unit is equivalent to the controller 613B-N. h The sub-model and the adaptive model applied to the sub-model. The adaptive model is the model shown in equations (1) to (6) above.

[0154] For example, data D21 indicates that head identification number 21, head 19B-21, is located on arm 15B-2, arm 15B-2 as... Figure 7 As shown, the head 19B-21 is the upper head among the two heads of the arm 15B-2. The controller 610B-21 that controls the head 19B-21 is equivalent to the controller 613B-21 of the feedforward control unit 63B, which consists of two sub-models F1 and F2.

[0155] Figure 9 (a) is one example; data items can be added or removed as needed. For instance, head 19B-N can be determined simply by having "arm position" and "head mounting position". h In some cases, the "head identification number N" may not be required. h That is, it is also possible to make it possible to simply put the head 19B-N h When either the "arm position" or "head mounting position" is entered into the database, the controller 613B-N is automatically associated. h The number and content of sub-models. In addition, regarding "arm position", although the top, middle and bottom three are indicated, it is also possible to further subdivide or roughly classify them for the vibration modes of each arm 15B.

[0156] The "adaptation model" can also be configured according to 19B-N for each head. h Based on considerations such as "arm position" and "head mounting position," one or more suitable sub-models are pre-selected and stored. Figure 9 In (a), an example is given where the overall vibration mode is divided into three sub-models: F1, F2, and F3. The vibration mode of each sub-model is determined in advance, and the transfer function model is modeled by formulas and parameters based on the vibration mode and stored in the database.

[0157] For example, the "adaptive model" of data D5 indicates that controller 613B-5 contains three sub-models F1, F2, and F3. If we assume controller 613B-5 is an adaptive model... Figure 5 If the controller 613A has the same configuration, then it can be implemented as shown in the second embodiment. Figure 6 The process involves identifying and updating the parameters of the transfer function models of sub-models F1, F2, and F3.

[0158] Each controller 613B-N h Retrieve the headers 19B-N of the controlled object from the database respectively. h The adaptive model, based on the acquired information, identifies or updates the parameters of the sub-model using methods as shown in embodiments 1 and 2. Additionally, each controller 613B-N... h Generate microactuators 17B-N respectively h MA control correction value (equivalent to) Figure 2 u_m2).

[0159] Each controller 610B-N h Based on target location r0N h and / or head position y0N h Controller 613B-N h The output MA control correction value is used to generate the VCM13B and microactuator 17B-N. hControl value (equivalent to) Figure 2 u_v, u_m1). VCM13B and microactuator 17B-N h Based on the input control value, control head 19-N h Determine the head position yN h (=y_m-N h +y_v-N h ).

[0160] For example, when the power supply of the disk drive 1B according to this embodiment is applied, each controller 613B-N of the feedforward control unit 63B... h For example, from non-volatile memory 90B and / or parameter update unit 64B-N h The latest parameters are obtained, and digital filter processing based on the adaptive model set according to the obtained parameters is performed. Additionally, the parameter update unit 64B-N... h Alternatively, based on the parameters obtained from the non-volatile memory 90B, further parameter updates can be performed, and the updated parameters can be provided to each controller 613B-N. h .

[0161] Furthermore, in the database, at least the head identification number N, which serves as head identification information, is... h An associative adaptation model is sufficient. For example, based separately on header 19B-N. h The "arm position" and "head mounting position" determine the "adaptation model," and only the "head identification number N" is stored in the database. h "Adaptive Model"

[0162] exist Figure 9 In the database example of (b), it indicates that the data is stored row-wise with header 19B-N. h Related information, according to each header 19B-N h Define a "group" for identifying groups, and specify one or more "adaptation models" based on the defined "groups". Although Figure 9 of (a), Figure 9 (b) is shown separately, but it can also be seen in Figure 9 Add a "group" column to (a) to treat it as the same database.

[0163] "Group" is, for example, based on each head 19B-N h The groups determined by the similarity of vibration modes and other grouping criteria can also have their values ​​stored in a database in advance based on the grouping criteria.

[0164] Figure 9 (b) indicates that the “group” is divided into 1 to 3 and each head is 19B-N h Specified examples. Grouping criteria, in addition to similarity of vibration modes, can also be the first 19B-N.h The arm position, head mounting position, etc., can be adjusted according to the number of heads 19B, etc.

[0165] Figure 9 The "adaptation model" in (b) can be associated with the "group" either automatically when the "group" is determined, or it can be specified independently of the "group." For example, Figure 9 Data D22 and D42 in (b) are examples of cases where the adaptation model is automatically set to F1 and F2 when the specified group is 2.

[0166] The "group" can be used to automatically specify the associated "adaptation model" as described above, or it can be associated with the parameter update frequency of the adaptation models contained in the "group". For example, parameter update unit 64B can be configured according to each header 19B-N. h The parameter update frequency of the associated "group" is updated in the controller 613B-N. h The parameters of the sub-model.

[0167] exist Figure 9 In the example of database (c), it indicates that the data is stored row-wise with header 19B-N. h Related information, and with Figure 9 Unlike case (b), all head 19B have three sub-model processing units (groups F1, F2, and F3). This example is grouped as follows: when viewed along the column direction (vertical direction)... Figure 9 In the case of database (c), group F1 contains sub-model F1 in common to all heads, group F2 contains two sub-models, F21 for the outer arm and F22 for the inner arm, and group F3 contains four sub-models, F31, F32, F33 and F34, from top to bottom, grouped into two sub-models.

[0168] Additionally, when observing along the column direction (longitudinal direction) Figure 9 In the case of database (c), for example Figure 9 Data D31 in (c) indicates that for the sub-model processing unit (F1 group) of head 19B-31, sub-model processing unit (F2 group) contains sub-model F22, and sub-model processing unit (F3 group) contains sub-model F32. Similarly, data D41 indicates that for the sub-model processing unit (F1 group) of head 19B-41, sub-model processing unit (F2 group) contains sub-model F22, and sub-model processing unit (F3 group) contains sub-model F33.

[0169] As shown above, it is possible to perform various groupings of head 19B, such as combining sub-model processing units and combining sub-models.

[0170] Through the above process, disk device 1B can determine the contents of multiple headers 19B-N.h The position of the arm, the mounting position of the head, etc., determine the multiple controllers 613B-N h The sub-model. Additionally, disk device 1B can be configured to store multiple headers 19B-N. h Groups obtained by grouping based on grouping criteria determine multiple controllers 613B-N. h The sub-model. Disk device 1B can control each head 19B-N. h 613B-N controller with more than one sub-model h Execute for each head 19B-N h Feedforward control.

[0171] According to at least one of the embodiments described above, a disk device and a head control method for suppressing head vibration through feedforward control can be provided.

[0172] Some embodiments of the present invention have been described, but these embodiments are provided as examples and are not intended to limit the scope of the invention. These new embodiments can be implemented in various other ways, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included within the scope and / or spirit of the invention, and are included within the scope of the claims and their equivalents. Furthermore, the processes shown in flowcharts, timing diagrams, etc., can also be implemented by hardware such as a CPU, IC chip, digital signal processor (DSP), or software (programs, etc.) operating in a computer, including a microcomputer, or a combination of hardware and software. Furthermore, the apparatus of the present invention can be applied when the claims are expressed as control logic, as a program containing instructions for causing a computer to execute, or as a computer-readable recording medium containing said instructions. Moreover, the names and terms 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 drive, comprising: The head reads data from the disk and writes data to the disk; Arm, supporting the head; The voice coil motor (VCM) controls the arm to perform seek control on the head; The micro-actuator, or MA, adjusts the position of the head; The servo control unit generates VCM control values ​​input to the VCM and MA control values ​​input to the microactuator; and The feedforward control unit receives the current position information of the head (i.e., head position information) and the VCM control value as input, and outputs a MA control correction value for the microactuator. The feedforward control unit includes a controlled object model processing unit that includes a transfer function model, wherein the transfer function model represents the relationship between the input values ​​input to the transfer function model and the output values ​​output from the transfer function model. The controlled object model processing unit also includes multiple sub-model processing units. Each of the sub-model processing units generates a sub-model output value based on the given transfer function sub-model. The sub-model output value is the output value for the input value input to the transfer function model. The feedforward control unit generates the MA control correction value based on the output values ​​of multiple sub-models.

2. The disk drive according to claim 1, At least one of the plurality of sub-model processing units includes adaptive parameters that can be updated in the given transfer function sub-model, and the feedforward control unit updates the adaptive parameters based on the VCM control value and the head position information.

3. The disk drive according to claim 2, The feedforward control unit includes: The first sub-model processing unit outputs the first sub-model output value and generates a first error value based on the first sub-model output value and the head position information; and The second sub-model processing unit outputs the second sub-model output value and generates a second error value based on the second sub-model output value and the first error value. The output value is generated based on the output value of the second sub-model. Either the first sub-model processing unit or the second sub-model processing unit updates the adaptive parameters based on the generated error value.

4. The disk drive according to claim 3, The feedforward control unit also includes multiple third sub-model processing units, which output third sub-model output values ​​and generate a third error value based on the third sub-model output values ​​and an error value received from another sub-model processing unit. At least one of the sub-model processing units updates the adaptive parameters based on the generated error values.

5. The disk drive according to any one of claims 1 to 4, The plurality of sub-model processing units include a second transfer function sub-model determined based on the vibration modes that affect the position of the head.

6. The disk drive according to claim 5, The vibration modes include the rigid body modes of the VCM.

7. The disk drive according to claim 6, The vibration mode includes the main resonance mode of the VCM.

8. The disk drive according to claim 5, The multiple sub-model processing units include a third transfer function sub-model defined by the ARX model.

9. The disk drive according to claim 8, It comprises: a plurality of said heads; a VCM for controlling the plurality of said heads; a plurality of said arms; a plurality of said microactuators; a plurality of said feedforward control units; a plurality of servo control units; and a storage unit for storing a first database. The first database stores head identification information as information for identifying the multiple heads, and a fourth transfer function sub-model associated with the head identification information and applicable to the sub-model. The plurality of feedforward control units determine the fifth transfer function sub-model of the sub-model processing unit based on the first database.

10. The disk drive according to claim 9, The head identification information is a combination of the mounting positions of each arm on which the head is mounted and the vertical position of the head.

11. The disk drive according to claim 10, The multiple heads are grouped according to a pre-defined grouping criterion. It possesses a second database, which contains group identification numbers associated with the plurality of heads and adaptation models associated with the group identification numbers. The feedforward control unit determines the sixth transfer function sub-model of the sub-model processing unit based on the second database.

12. A head control method for a disk device, The disk device includes: The head reads data from the disk and writes data to the disk; Arm, supporting the head; The voice coil motor (VCM) controls the arm to perform seek control on the head; The micro-actuator, or MA, adjusts the position of the head; The servo control unit generates VCM control values ​​input to the VCM and MA control values ​​input to the microactuator; and The feedforward control unit receives the current position information of the head (i.e., head position information) and the VCM control value as input, and outputs a MA control correction value for the microactuator. The head control method includes: The feedforward control unit is equipped with a controlled object model that includes a transfer function model, wherein the transfer function model represents the relationship between the input values ​​input to the transfer function model and the output values ​​output from the transfer function model. The controlled object model is further decomposed into multiple sub-models. Each of the sub-models is given a transfer function sub-model, and each sub-model generates a sub-model output value based on the given transfer function sub-model. The sub-model output value is the output value for the input value input to the transfer function model. The MA control correction value is generated based on the output values ​​of multiple sub-models.

Images