Magnetic disk device

Abstract

To provide a magnetic disk device having high positioning accuracy.SOLUTION: A controller of a magnetic disk device acquires a first amplitude and a first phase of a first read waveform and a second amplitude and a second phase of a second read waveform by demodulating the first read waveform which is a read waveform of a first burst pattern and the second read waveform which is a read waveform of a second burst pattern. The controller acquires weights based on the first amplitude, the second amplitude, the first phase, and the second phase. The controller corrects the initial phases of the first and second lead waveforms by calculating weighted averages using the weights, the first phase, and the second phase, and performs positioning of the magnetic head based on the corrected first and second lead waveforms.SELECTED DRAWING: Figure 8

Description

[Technical field]

[0001] The present embodiment relates to a magnetic disk device. [Background technology]

[0002] In magnetic disk drives, burst patterns used for positioning the magnetic head are recorded in servo sectors on the magnetic disk. In recent years, in order to realize larger capacity magnetic disk drives, there are magnetic disk drives in which one or more short servo sectors are arranged between two regular servo sectors (hereinafter referred to as normal servo sectors) in the circumferential direction. A short servo sector is a servo sector that is shorter in length in the circumferential direction than a normal servo sector.

[0003] In addition to the burst pattern, data called servo marks are recorded in the normal servo sector. In the normal servo sector, a burst gate indicating the timing to read the burst pattern is generated based on the servo mark. However, in the short servo sector, the burst gate is generated without using the servo mark. Therefore, in the short servo, the read timing of the burst pattern deviates from the proper timing, which may result in a deterioration in positioning accuracy. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] U.S. Pat. No. 1,074,8569 Summary of the Invention [Problem to be solved by the invention]

[0005] An object of one embodiment is to provide a magnetic disk drive with high positioning accuracy. [Means for solving the problem]

[0006] According to one embodiment, a magnetic disk device includes a magnetic disk, a magnetic head, and a controller. The magnetic disk has a plurality of servo sectors arranged in a circumferential direction, and a first burst pattern and a second burst pattern are recorded in each of the plurality of servo sectors. The magnetic head performs writing and reading on the magnetic disk. The controller performs the following operation when the magnetic head passes over a first servo sector that is one of the plurality of servo sectors. That is, the controller specifies the read timing of the first burst pattern and the second burst pattern, and reads the first burst pattern and the second burst pattern from the first servo sector at the specified read timing. The controller obtains a first amplitude and a first phase of the first read waveform and a second amplitude and a second phase of the second read waveform by demodulating a first read waveform that is a read waveform of the first burst pattern and a second read waveform that is a read waveform of the second burst pattern. The controller obtains weights based on the first amplitude, the second amplitude, the first phase, and the second phase. The controller corrects the initial phases of the first read waveform and the second read waveform based on an initial phase correction value obtained by calculating a weighted average using the weight, the first phase, and the second phase. The controller performs positioning of the magnetic head based on the first read waveform and the second read waveform after the correction of the initial phase. [Brief description of the drawings]

[0007] [Figure 1] FIG. 1 is a schematic diagram illustrating an example of the configuration of a magnetic disk device according to the first embodiment. [Diagram 2] FIG. 2 is a schematic diagram showing an example of the configuration of the magnetic disk according to the first embodiment. [Diagram 3] FIG. 3 is a diagram showing an example of the configuration of a normal servo sector and a short servo sector according to the first embodiment. [Figure 4] FIG. 4 is a diagram showing an example of the configuration of an N burst NB and a Q burst QB according to the first embodiment. [Diagram 5] FIG. 5 is a diagram for explaining the initial phase operation according to the first embodiment. [Figure 6] FIG. 6 is a diagram showing an example of the relationship between the offset amount from the track center and the absolute value of the phase difference. [Figure 7] FIG. 7 is a diagram illustrating an example of a method for determining weights according to the first embodiment. [Figure 8] FIG. 8 is a flowchart showing an example of the positioning control operation according to the first embodiment. [Figure 9] FIG. 9 is a diagram showing the initial phase obtained by the series of operations shown in FIG. [Figure 10] FIG. 10 is a diagram for explaining an example of a method for selecting a plurality of weighting tables according to the second embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of the configuration of a plurality of weighting tables according to the second embodiment. [Figure 12] FIG. 12 is a flowchart showing an example of the positioning control operation according to the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, a magnetic disk device according to an embodiment will be described in detail with reference to the accompanying drawings. Note that the present invention is not limited to these embodiments.

[0009] (First embodiment) FIG. 1 is a schematic diagram showing an example of the configuration of a magnetic disk device 1 according to the first embodiment.

[0010] The magnetic disk device 1 is connected to a host 2. The magnetic disk device 1 can receive access commands from the host 2. The access commands are, for example, write commands and read commands.

[0011] The magnetic disk device 1 includes a magnetic disk 11 having a magnetic layer formed on its surface. The magnetic disk device 1 accesses the magnetic disk 11 in response to an access command. The access includes writing data and reading data. In detail, writing and reading data is performed by a magnetic head 22.

[0012] In addition to the magnetic disk 11, the magnetic disk device 1 includes a spindle motor (SPM) 12, a ramp 13, an actuator arm 15, a voice coil motor (VCM) 16, a servo controller (SVC) 21, a magnetic head 22, a hard disk controller (HDC) 23, a preamplifier 24, a read / write channel (RWC) 25, a processor 26, a Flash Read Only Memory (FROM) 28, and a Dynamic Random Access Memory (DRAM) 29.

[0013] The magnetic disk 11 is rotated at a predetermined rotational speed by the SPM 12 attached coaxially.

[0014] The SVC 21 is an integrated circuit having a function as a driver for driving the SPM 12 and the VCM 6. The processor 26 controls the rotation of the SPM 12 and the VCM 16 via the SVC 21.

[0015] The magnetic head 22 writes and reads data to and from the magnetic disk 11 by means of a write head 22w and a read head 22r provided thereto. The magnetic head 22 is attached to the tip of an actuator arm 15. The magnetic head 22 is moved in the radial direction of the magnetic disk 11 by a VCM 16 driven by an SVC 21. Note that a single magnetic head 22 may be provided with a plurality of either one of the write heads 22w and / or the read heads 22r.

[0016] When the magnetic disk 11 is stopped rotating, the magnetic head 22 is moved onto the ramp 13. The ramp 13 holds the magnetic head 22 at a position separated from the magnetic disk 11.

[0017] The preamplifier 24 is an integrated circuit that writes and reads data via the magnetic head 22. During a read operation, the preamplifier 24 amplifies and outputs a signal read by the magnetic head 22 from the magnetic disk 11, and supplies the signal to the RWC 25. During a write operation, the preamplifier 24 amplifies a signal corresponding to the data to be written, which is supplied from the RWC 25, and supplies the signal to the magnetic head 22.

[0018] The HDC 23 controls data transmission and reception between the host 2 via the I / F bus, and controls the DRAM 29 .

[0019] The DRAM 29 is used as a buffer for data transmitted to and received from the host 2. For example, the DRAM 29 is used to temporarily store data to be written or data read from the magnetic disk 11.

[0020] Furthermore, the DRAM 29 is used as an operating memory by the processor 26. The DRAM 29 is used as an area into which firmware programs are loaded and an area in which various types of management data are temporarily stored.

[0021] The RWC 25 modulates the data to be written that is supplied from the HDC 23 and supplies the modulated data to the preamplifier 24. The RWC 25 also performs demodulation, including error correction, on the signal that is read from the magnetic disk 11 and supplied from the preamplifier 24, and then outputs the signal to the HDC 23 as digital data.

[0022] The processor 26 is, for example, a CPU (Central Processing Unit). A FROM (Flash Read Only Memory) 28 and a DRAM 29 are connected to the processor 26.

[0023] The FROM 28 stores firmware programs and various setting information.

[0024] The processor 26 performs overall control of the magnetic disk device 1 in accordance with a firmware program stored in the FROM 28 or the magnetic disk 11. For example, the processor 26 loads the firmware program from the FROM 28 or the magnetic disk 11 into the DRAM 29, and controls the SVC 21, the preamplifier 24, the RWC 25, the HDC 23, and the like in accordance with the firmware program loaded into the DRAM 29.

[0025] The firmware program may be stored in the magnetic disk 11. Some or all of the functions of the processor 26 may be realized by a hardware circuit such as a field-programmable gate array (FPGA) or an application specific integrated circuit (ASIC).

[0026] The HDC 23, the RWC 25, and the processor 26 are configured as a single integrated circuit, a System-On-a-Chip (SoC) 30. The SoC 30 may include other elements (such as a ROM 28 or a DRAM 29) in addition to the above. The SoC 30 is an example of a controller.

[0027] 2 is a schematic diagram showing an example of the configuration of the magnetic disk 11 according to the first embodiment. Note that this diagram shows an example of the rotation direction of the magnetic disk 11. The magnetic head 22 moves relative to the magnetic disk 11 as the magnetic disk 11 rotates. Therefore, the write / read direction, i.e., the direction in which data is written or read by the magnetic head 22 along the circumferential direction, is opposite to the rotation direction of the magnetic disk 11.

[0028] In the radial direction, the direction from the edge to the center of the magnetic disk 11 is the inner diameter (ID) direction, and the direction from the center to the edge of the magnetic disk 11 is the outer diameter (OD) direction.

[0029] In the manufacturing process, servo data used for positioning the magnetic head 22 is written on the magnetic disk 11, for example, by a servo writer or by self-servo writing (SSW). As shown in Fig. 2, as an example of the arrangement of servo areas in which servo data is written, a plurality of servo areas SV arranged radially in the radial direction and at a predetermined interval in the circumferential direction are depicted. The space between two consecutive servo areas SV in the circumferential direction is used as a data area DA in which data is written.

[0030] The multiple servo areas SV include normal servo areas NSV and short servo areas SSV. At least one short servo area SSV is arranged between two normal servo areas NSV arranged consecutively in the circumferential direction, in other words, between two adjacent normal servo areas NSV. According to the example shown in FIG. 2, the normal servo areas NSV and the short servo areas SSV are arranged alternately in the circumferential direction.

[0031] A plurality of concentric servo tracks 41 are provided in the radial direction of the magnetic disk 11. The servo data written in the servo areas SV is used to control the positioning of the magnetic head 22.

[0032] More specifically, a plurality of concentric data tracks are provided on an area of ​​the magnetic disk 11 where the plurality of servo tracks 41 are provided. The plurality of servo tracks 41 may be used as a plurality of data tracks, or a plurality of data tracks different from the plurality of servo tracks 41 may be provided. A plurality of data tracks continuous in the circumferential direction are provided in an area divided by the data areas DA on each data track. Data can be written to each data track. The data that can be written to each data track includes user data received from the host 2, metadata (e.g., error correction code) associated with the user data, system data, and the like. The magnetic disk device 1 holds in advance the setting of the positional relationship between the plurality of servo tracks 41 and the plurality of data tracks, and performs positioning control to position the magnetic head 22 on a target data track based on the servo data recorded in the servo areas SV. The positioning control includes a seek operation, which is an operation of moving the magnetic head 22 in the radial direction toward the target data track, and a tracking operation to keep the magnetic head 22 on the target data track.

[0033] Hereinafter, an area on the servo track 41 divided by the normal servo areas NSV will be referred to as a normal servo sector NSV. An area on the servo track 41 divided by the short servo areas SSV will be referred to as a short servo sector SSV. Moreover, the normal servo sectors NSV and the short servo sectors SSV will be collectively referred to as a servo sector SV.

[0034] In this specification, the positional relationship between two regions in the circumferential direction is defined. When first data and second data are written along the write / read direction, and the first data is written in an area that is read before the second data, the region where the first data is written is considered to be located before the region where the second data is written. The region where the second data is written is considered to be located after the region where the first data is written.

[0035] In addition, when the magnetic head 22 passes over a certain region in the circumferential direction, the part of the region that the magnetic head 22 passes over first is considered to be the beginning of the region, and when the magnetic head 22 passes over a certain region in the circumferential direction, the part of the region that the magnetic head 22 passes over last is considered to be the end of the region.

[0036] FIG. 3 is a diagram showing an example of the configuration of the normal servo sector NSV and the short servo sector SSV according to the first embodiment.

[0037] All normal servo sectors NSV have the same configuration. Also, all short servo sectors SSV have the same configuration. In FIG. 3, the configuration of servo sector SV_i, which is a normal servo sector NSV, is shown as a representative of normal servo sectors NSV. Also, the configuration of servo sector SV_i+1, which is arranged between servo sector SV_i, which is a normal servo sector NSV, and servo sector SV_i+2, which is a normal servo sector NSV, is shown as a representative of short servo sectors SSV.

[0038] As shown in FIG. 3, in the normal servo sector NSV, a preamble PA, a servo mark SM, a gray code GC, an N burst NB, a Q burst QB, and a post code PC are recorded in this order along the write / read direction.

[0039] The preamble PA is a single periodic pattern data that changes periodically in the circumferential direction. The preamble PA is used to adjust the amplitude, phase and frequency of the sampling data when the servo waveform read by the read head 22r is taken into the RWC 25 as sampling data based on the servo clock. The servo clock is generated by the RWC 25.

[0040] The servo mark SM is pattern data for specifying the read timing of the servo data. The SoC 30 specifies the read timing of various servo data based on the detection timing of the servo mark SM and the count value of a counter (not shown) included in the SoC 30. An example of specifying the read timing of various servo data based on the detection timing of the servo mark SM will be described later.

[0041] The Gray code GC includes a cylinder address for identifying each servo track 41 provided on the magnetic disk 11 and a sector address for identifying each servo sector SV on the servo track 41 .

[0042] The N burst NB and the Q burst QB are pattern data used to detect the offset amount of the position of the magnetic head 22 from the track center of a certain servo track 41 (more precisely, the servo track 41 indicated by the cylinder address). Each of the N burst NB and the Q burst QB is also referred to as a burst pattern. The N burst NB is an example of a first burst pattern. The Q burst QB is an example of a second burst pattern.

[0043] Each burst pattern is captured by the RWC 25 at a sampling interval based on the servo clock. The RWC 25 performs, for example, a Discrete Fourier Transform (DFT) process on the waveform of each captured burst pattern as one of the demodulation operations, thereby acquiring the phase and amplitude. The SoC 30 (e.g., the processor 26) calculates the offset amount of the magnetic head 22 from the track center of the servo track 41 based on the phase and amplitude acquired by the RWC 25. The SoC 30 (e.g., the processor 26) calculates an estimated radial position of the magnetic head 22 based on the cylinder address obtained from the Gray code GC and the offset amount obtained from each burst pattern.

[0044] Hereinafter, the waveform of the burst pattern captured by the RWC 25 will be referred to as the read waveform. Also, the offset amount from the track center of the position of the magnetic head 22 detected based on the burst pattern will be referred to as the burst offset. Also, the estimated position of the magnetic head 22 in the radial direction will be referred to as the demodulation position.

[0045] The postcode PC is data for correcting the positional deviation of the shape of the servo track 41, which is defined by the gray code GC, the N burst NB, and the Q burst QB, from the ideal shape of the servo track 41. The degree of this positional deviation varies in synchronization with the rotation of the magnetic disk 11. Therefore, the positional deviation is also called RRO (Repeatable RunOut). That is, the postcode PC is data for correcting the RRO. The SoC 30 corrects the demodulation position using the postcode PC.

[0046] FIG. 3 further illustrates an example of the waveforms of the servo gate signal SG, the burst gate signal BG, and the servo mark found signal SMF.

[0047] The servo gate signal SG is a signal that the processor 26 sends to the RWC 25, and indicates a period during which reading of servo data is permitted. In the example shown in Fig. 3, when the servo gate signal SG is at "H" level, it indicates that reading of servo data is permitted, and when the servo gate signal SG is at "L" level, it indicates that reading of servo data is not permitted. The RWC 25 executes reading of servo data during the period during which the servo gate signal SG is at "H" level.

[0048] The burst gate signal BG is an internal signal of the RWC 25. The burst gate signal BG is a signal that indicates the period during which the read of the burst pattern (i.e., the N burst NB and the Q burst QB) is executed. In the example shown in Fig. 3, the timing when the burst gate signal BG rises from the "L" level to the "H" level indicates the timing when the read of the burst pattern starts, and the timing when the burst gate signal BG falls from the "H" level to the "L" level indicates the timing when the read of the burst pattern ends. In other words, the period during which the burst gate signal BG is maintained at "H" indicates the read period of the burst pattern.

[0049] The servo mark found signal SMF is a signal indicating whether or not the servo mark SM has been correctly detected. The servo mark found signal SMF is controlled by, for example, the RWC 25 or the processor 26. In the example shown in Fig. 3, the servo mark found signal SMF is basically maintained at "H" level. Then, when the magnetic head 22 passes over an area in which the servo mark SM is recorded, if the servo mark SM is correctly detected, the servo mark found signal SMF is lowered to "L" level and then immediately raised to "H" level.

[0050] The SoC 30 is configured to operate the burst gate signal BG based on the timing at which the servo mark found signal SMF rises when the magnetic head 22 passes over the normal servo sector NSV.

[0051] The positional relationship between the area where each burst pattern is recorded and the area where the servo marks SM are recorded, and the rotation speed of the magnetic disk 11 are known. Therefore, the relationship between the timing when the magnetic head 22 passes over the area where the servo marks SM are recorded and the timing when the magnetic head 22 passes over the area where each burst pattern is recorded is known. Based on the relationship between these timings, the rising and falling timings of the burst gate signal BG are determined.

[0052] In detail, when the SoC 30 correctly detects the servo mark SM in the normal servo sector NSV, the SoC 30 drops the servo mark found signal SMF and then immediately raises it (timing t0). Then, at timing t1, which is a predetermined delay time dn after timing t0, the RWC 25 starts an operation of raising the burst gate signal BG, maintaining the burst gate signal BG at the “H” level for a predetermined period, and lowering the burst gate signal BG.

[0053] The operation of raising the burst gate signal BG, maintaining the burst gate signal BG at the "H" level for a predetermined period, and lowering the burst gate signal BG is performed for each of the N burst NB and the Q burst QB. The timing of the rising edge of the burst gate signal BG and the timing of the falling edge of the burst gate signal BG are set so that the period during which the burst gate signal BG is maintained at the "H" level the first time allows the read head 22r to read a predetermined partial area of ​​the area in which the N burst NB is recorded. The timing of the rising edge of the burst gate signal BG and the timing of the falling edge of the burst gate signal BG are set so that the period during which the burst gate signal BG is maintained at the "H" level the second time allows the read head 22r to read a predetermined partial area of ​​the area in which the Q burst QB is recorded.

[0054] Therefore, in the normal servo sector NSV, RWC25 can read a specified portion of the N burst NB during the period in which the burst gate signal BG is maintained at the "H" level the first time, and can read a specified portion of the Q burst QB during the period in which the burst gate signal BG is maintained at the "H" level the second time.

[0055] In the short servo sector SSV, N burst NB and Q burst QB are recorded in this order along the write / read direction. In other words, the short servo sector SSV has a configuration in which the areas in which the preamble PA, servo mark SM, gray code GC, and post code PC are recorded are omitted from the normal servo sector NSV. Therefore, the short servo sector SSV has a shorter circumferential length than the normal servo sector NSV, and as a result, it is possible to increase the amount of user data that can be recorded in the data area DA.

[0056] During the period when the read head 22r passes over the short servo sector SSV, the servo gate signal SG is maintained at the "H" level. More specifically, the servo gate signal SG is operated so that the servo gate signal SG is maintained at the "H" level during the period including the period when the magnetic head 22 passes over the area where the N burst NB is recorded and the area where the Q burst QB is recorded. The RWC 25 operates the burst gate signal BG during the period when the servo gate signal SG is maintained at the "H" level. Then, the RWC 25 executes demodulation on the burst pattern read during the period defined by the burst gate signal BG.

[0057] No servo mark SM is recorded in the short servo sector SSV. Therefore, the RWC 25 determines the transition timing of the burst gate signal BG based on the timing at which the servo mark SM is detected when the magnetic head 22 passes over another servo sector SV corresponding to the normal servo sector NSV.

[0058] In detail, at timing t2, which is a predetermined delay time ds after timing t0 when the servo mark found signal SMF is raised in the servo sector SV_i, which is the normal servo sector NSV located immediately before, RWC25 starts the operation of raising the burst gate signal BG, maintaining the burst gate signal BG at the “H” level for a predetermined period of time, and lowering the burst gate signal BG.

[0059] The operation of raising the burst gate signal BG, maintaining the burst gate signal BG at the "H" level for a predetermined period, and lowering the burst gate signal BG is executed for each of the N burst NB and the Q burst QB, similar to when reading each burst pattern from the normal servo sector NSV. Therefore, even in the short servo sector SSV, the RWC25 can read a predetermined portion of the N burst NB during the period when the burst gate signal BG is maintained at the "H" level the first time, and can read a predetermined portion of the Q burst QB during the period when the burst gate signal BG is maintained at the "H" level the second time.

[0060] The configuration of the short servo sector SSV shown in Fig. 3 is merely an example. For example, the short servo sector SSV may have a preamble PA, a servo mark SM, and a gray code GC recorded before the area where the N burst NB is recorded. However, in the positioning control during a specific operation (for example, the operation of writing data to the data area DA), as shown in Fig. 3, when the read head 22r passes over the short servo sector SSV, the reading of servo data is started from the N burst NB, and the servo data recorded before the area where the N burst NB is recorded is not read. This makes it possible to increase the amount of user data that can be recorded in the data area DA.

[0061] 4 is a diagram showing an example of the configuration of an N burst NB and a Q burst QB according to the first embodiment. n and servo track TR n The servo track TR adjacent to n+1 The diagram shows an example of the configuration of the N burst NB and the Q burst QB around the . The unit patterns in black indicate one of two values ​​with different polarities (e.g., magnetization with a positive polarity), and the unit patterns in white indicate the other of the two values ​​(e.g., magnetization with a negative polarity). Hereinafter, the unit patterns in black will be referred to as the first unit pattern, and the unit patterns in white will be referred to as the second unit pattern.

[0062] The length in the radial direction may be expressed in servo track units, in other words, units of spatial length assuming the servo track pitch to be 1. The unit of spatial length assuming the servo track pitch to be 1 is denoted as Svtrk.

[0063] In the example shown in Fig. 4, each of the N burst NB and the Q burst QB has a configuration in which the first unit pattern and the second unit pattern are alternately arranged in the radial and circumferential directions, respectively. In the radial direction, each unit pattern has a length of 1Svtrk.

[0064] The N burst NB is recorded so that the boundary between the first unit pattern and the second unit pattern coincides with the track center. Therefore, when the magnetic head 22 passes the track center, the amplitude of the read waveform of the N burst NB becomes extremely small (ideally zero). Also, when the magnetic head 22 passes a position offset by 0.5 Svtrk from the track center, the amplitude of the read waveform of the N burst NB becomes extremely large.

[0065] Hereinafter, the boundary between the first unit pattern and the second unit pattern in the radial direction will be referred to as a pattern boundary.

[0066] The Q burst QB is recorded so that the pattern boundary coincides with a position offset by 0.5Srvtrk from the track center. Therefore, when the magnetic head 22 passes the track center, the amplitude of the read waveform of the Q burst QB becomes maximum. Also, when the magnetic head 22 passes a position offset by 0.5Svtrk from the track center, the amplitude of the read waveform of the Q burst QB becomes minimum (ideally zero).

[0067] The partial area PN is a regular read portion of the area where the N burst NB is recorded. The partial area PQ is a regular read portion of the area where the Q burst QB is recorded. The SoC 30 is configured to obtain a correct burst offset by demodulating the read waveform of the N burst NB read from the partial area PN and the read waveform of the Q burst QB read from the partial area PQ. Therefore, the delay times dn, ds and various set times are set so that the burst gate signal BG can rise when the magnetic head 22 reaches the top of the partial area PN, fall when the magnetic head 22 reaches the end of the partial area PN, rise when the magnetic head 22 reaches the top of the partial area PQ, and fall when the magnetic head 22 reaches the end of the partial area PQ.

[0068] Here, the delay time ds is significantly longer than the delay time dn. Therefore, in the short servo sector SSV, the read timing of each burst pattern is likely to deviate from the normal timing, that is, the timing at which the partial area PN and the partial area PQ can be read, due to the influence of the rotation fluctuation of the magnetic disk 11. Even if the RWC 25 raises the burst gate signal BG at the timing when the delay time dn has elapsed from the timing t0, for example, if the rotation fluctuation of the magnetic disk 11 occurs during the period from the timing t0 to the delay time dn, the timing at which the burst gate signal BG rises may deviate from the timing at which the magnetic head 22 reaches the top of the partial area PN or the partial area PQ. That is, the read timing of each burst pattern may deviate from the normal timing.

[0069] If the read timing of each burst pattern deviates from the normal timing, the demodulation result of each burst pattern fluctuates, and the demodulation position calculated based on the demodulation result of each burst pattern may be incorrect. If the incorrect demodulation position is used for positioning control, the accuracy of the positioning control will deteriorate.

[0070] 4, the same read waveform is observed when the position of the magnetic head 22 is shifted by 1Svtrk in the radial direction and when the read timing of the burst pattern defined by the burst gate signal BG is shifted by the time equivalent to one unit pattern (i.e., the first unit pattern or the second unit pattern). Therefore, it is difficult to judge whether the burst offset obtained based on the read waveform is an incorrect burst offset caused by a shift in the read timing of the burst pattern or whether the position of the magnetic head 22 is really shifted in the radial direction.

[0071] Therefore, the SoC 30 is configured to execute an operation to correct the initial phase when the phase obtained by performing DFT processing on the read waveform (hereinafter, referred to as the initial phase) shifts due to a shift in the read timing of the burst pattern. This operation is referred to as an initial phase correction operation.

[0072] FIG. 5 is a diagram for explaining the initial phase operation according to the first embodiment.

[0073] Part (A) of Fig. 5 is a graph showing the relationship between the offset from the track center and the amplitude obtained by performing DFT processing on the read waveform of each burst pattern. The amplitude obtained from the read waveform of N burst NB is denoted as Namp. The amplitude obtained from the read waveform of Q burst QB is denoted as Qamp.

[0074] As shown in part (A), the amplitude Namp is minimum around each track center and maximum around each position offset by 0.5 Svtrk from the track center. The amplitude Qamp is minimum around each position offset by 0.5 Svtrk from the track center and maximum around each track center.

[0075] Part (B) of Fig. 5 is a graph showing the relationship between the offset from the track center and the phase obtained by performing DFT processing on the read waveform of each burst pattern. The phase obtained from the read waveform of N burst NB is denoted as Nphs. The phase obtained from the read waveform of Q burst QB is denoted as Qphs.

[0076] The SoC 30 corrects the phases Nphs and Qphs so that the difference between the phases Nphs and Qphs is within 90 degrees with respect to the phases shown in part (B) of Fig. 5. This correction is referred to as a folding operation. The folding operation will be described in detail later.

[0077] Part (C) of FIG. 5 is a graph showing the relationship between the offset amount from the track center and the phases Nphs and Qphs after the turn-back operation.

[0078] As shown in part (C), the phases Nphs and Qphs are close to the initial phase at the positions where the amplitude is maximum. However, at positions where a large amplitude is not obtained, particularly at pattern boundaries, the phases Nphs and Qphs are significantly different from the initial phase.

[0079] In order to calculate the initial phase from the phase Nphs and the phase Qphs, the SoC 30 first calculates an amplitude squared ratio, which is the ratio between the value obtained by squaring the amplitude Namp and the value obtained by squaring the amplitude Qamp. As an example, the ratio Rn of the N burst NB obtained by the following formula (1) is applied as the amplitude squared ratio. Rn=Namp 2 / (Namp 2 +Qamp 2 ) ···(1)

[0080] The ratio Rn of the N burst NB and the ratio Rq of the Q burst QB have the relationship shown in Equation (2). The ratio Rq of the Q burst QB may be used as the amplitude squared ratio in the subsequent processing. Rn+Rq=1 (2)

[0081] Part (D) of Fig. 5 is a graph showing the relationship between the offset amount from the track center of a certain servo track 41 and the amplitude squared ratio Rn. As shown in this figure, the amplitude squared ratio Rn has a value close to 0 at the pattern boundary of N burst NB, and a value close to 1 at the pattern boundary of Q burst QB. The amplitude squared ratio Rn changes approximately linearly with respect to the offset amount between the pattern boundary of N burst NB and the pattern boundary of Q burst QB.

[0082] The SoC 30 calculates a weight Nwt to be applied to the phase Nphs and a weight Qwt to be applied to the phase Qphs based on the magnitude squared ratio Rn, and obtains an initial phase correction value by calculating a weighted average using the weights Nwt, Qwt and the phases Nphs, Qphs.

[0083] The initial phase correction value is a value that is subtracted from the phase to set the initial phase to 0 degrees. In the example shown in Fig. 5, it can be seen that the initial phase is approximately minus 60 degrees from the phase Nphs and phase Qphs at the position where the amplitude is maximum. When the initial phase is minus 60 degrees, the initial phase correction value is minus 60 degrees.

[0084] A technique to be compared with the embodiment will be described below. The technique to be compared with the embodiment will be referred to as a comparative example.

[0085] According to the comparative example, the relationship between the amplitude square ratio Rn and the weights Nwt and Qwt is defined by a weight table. The weight table according to the comparative example is configured such that when the amplitude square ratio Rn is 0, the weight Nwt is set to 0, and when the amplitude square ratio Rn is 1, the weight Nwt is set to 1, and the closer the amplitude square ratio Rn is to 0, the closer the weight Nwt is to 0, and the closer the amplitude square ratio Rn is to 1, the closer the weight Nwt is to 1. The weight Qwt is calculated by the following formula (3). Qwt = 1 - Nwt (3)

[0086] 5, in each of the N burst NB and the Q burst QB, the closer the magnetic head 22 is to the pattern boundary, the more the phase of the read waveform deviates from the actual initial phase. According to the comparative example, the weight Nwt is made smaller as the magnetic head 22 is closer to the pattern boundary of the N burst NB, and the weight Qwt is made smaller as the magnetic head 22 is closer to the pattern boundary of the Q burst QB. Therefore, it is possible to suppress the influence of a phase that is significantly different from the actual initial phase obtained at the pattern boundary on the calculated value of the initial phase.

[0087] As described above, ideally, the amplitude Namp is 0 at the pattern boundary of the N burst NB, and the amplitude Qamp is 0 at the pattern boundary of the Q burst QB. However, in reality, the pattern boundary may become unclear due to poor writing quality of each unit pattern (first unit pattern and second unit pattern), and this may cause the amplitude Namp to be greater than 0 at the pattern boundary of the N burst NB, and the amplitude Qamp to be greater than 0 at the pattern boundary of the Q burst QB.

[0088] 5, the amplitude at the pattern boundary, i.e., the minimum value of the amplitude, is distributed in a range 100 greater than 0. In the comparative example, when the amplitude that should ideally be 0 at the pattern boundary takes a value significantly greater than 0, the amplitude squared ratio Rn takes a value deviating from 0 or 1. As a result, a weight somewhat greater than 0 is applied to a phase that deviates significantly from the actual initial phase, and the calculated value of the initial phase is influenced by the phase that deviates significantly from the actual initial phase, deteriorating the accuracy of the calculation of the initial phase.

[0089] 5, the position L1 where the offset amount is minus 0.5Svtrk corresponds to the pattern boundary of the Q burst QB. At this position L1, the amplitude Qamp reaches the maximum value in the range 100 in which the minimum amplitude values ​​are distributed, and as a result, the amplitude squared ratio Rn, which should ideally be 1, takes a value that is significantly different from 1.

[0090] As described above, the weight table of the comparative example is configured such that when the amplitude square ratio Rn is 0, the weight Nwt is 0, when the amplitude square ratio Rn is 1, the weight Nwt is 1, the closer the amplitude square ratio Rn is to 0, the closer the weight Nwt is to 0, and the closer the amplitude square ratio Rn is to 1, the closer the weight Nwt is to 1. Therefore, at the position L1, since the amplitude square ratio Rn is deviated from 1, the weight Nwt is deviated from 1, and the weight Qwt is deviated from 0. Ideally, at the position L1 where the weight Qwt should be 0, the weight Qwt is deviated from 0, so that a weight Qwt larger than 0 is applied to the phase Qphs that is greatly deviated from the actual initial phase. As a result, the calculated value of the initial phase is affected by the phase Qphs, and the accuracy of the calculation of the initial phase is deteriorated.

[0091] Thus, according to the comparative example, the accuracy of the calculated value of the initial phase may deteriorate when the magnetic head 22 passes over the pattern boundary, which may result in a deterioration in the accuracy of the positioning control.

[0092] Therefore, in the embodiment, the SoC 30 is configured to obtain the weights Nwt and Qwt using not only the magnitude squared ratio Rn but also the difference between the phase Nphs and the phase Qphs.

[0093] A method for obtaining the weights Nwt and Qwt based on the difference between the phase Nphs and the phase Qphs will be described with reference to Figures 6 and 7. Hereinafter, the absolute value of the difference between the phase Nphs and the phase Qphs will be referred to as the absolute value of the phase difference.

[0094] 6 is a diagram showing an example of the relationship between the offset amount from the track center of a certain servo track 41 and the absolute value of the phase difference. As shown in this diagram, the absolute value of the phase difference reaches a maximum value of 90 degrees at each pattern boundary of the N burst NB (i.e., each track center) and each pattern boundary of the Q burst QB (i.e., each position offset by 0.5 Svtrk from the track center). Between the pattern boundaries, the absolute value of the phase difference becomes smaller depending on the distance from the pattern boundary.

[0095] A certain value Th0 smaller than 90 degrees is set as a threshold value in SoC30. SoC30 compares the absolute value of the phase difference with the threshold value Th0. The absolute value of the phase difference being larger than the threshold value Th0 means that the magnetic head 22 has passed over a certain pattern boundary. However, depending on the pattern quality, the amplitude square ratio Rn may deviate from 0 or 1. The absolute value of the phase difference being smaller than the threshold value Th0 means that the magnetic head 22 has passed over a location away from the pattern boundary, and that the magnetic head 22 is located at a location where the amplitude square ratio Rn deviates from 0 or 1, in other words, where the amplitude square ratio Rn is close to 0.5. Based on the result of the comparison between the absolute value of the phase difference and the threshold value Th0, SoC30 determines whether or not to apply a weight table. The threshold value Th0 is an example of a first threshold value.

[0096] 7 is a diagram showing an example of a method for determining weights according to the first embodiment. In the example shown in this figure, the weight table Wtable is configured so that when the amplitude square ratio Rn is 0, the weight Nwt is 0, when the amplitude square ratio Rn is 1, the weight Nwt is 1, the closer the amplitude square ratio Rn is to 0, the closer the weight Nwt is to 0, and the closer the amplitude square ratio Rn is to 1, the closer the weight Nwt is to 1. The weight table Wtable may or may not be the same as the weight table according to the comparative example.

[0097] If the absolute value of the phase difference is smaller than the threshold value Th0, the SoC 30 obtains the weights Nwt and Qwt using the weight table Wtable.

[0098] When the absolute value of the phase difference is greater than the threshold value Th0, the SoC 30 sets each of the weights Nwt and Qwt to either 0 or 1.

[0099] Relationship 200 shows the relationship between the amplitude squared ratio Rn and the weight Nwt when the amplitude squared ratio Rn is greater than 0.5. As shown in relationship 200, when the amplitude squared ratio Rn is greater than 0.5, SoC30 sets the weight Nwt to 1 and the weight Qwt to 0. When the amplitude squared ratio Rn is less than 0.5, SoC30 sets the weight Nwt to 0 and the weight Qwt to 1.

[0100] That is, when it is estimated that the magnetic head 22 has passed over the pattern boundary of the N burst NB, the SoC 30 sets the weight Nwt to 0, and when it is estimated that the magnetic head 22 has passed over the pattern boundary of the Q burst QB, the weight Qwt to 0. This makes it possible to completely suppress the influence of a phase that is significantly deviated from the initial phase on the calculated value of the initial phase at the pattern boundary, and to suppress deterioration of the accuracy of the calculation of the initial phase.

[0101] When the absolute value of the phase difference is equal to the threshold value Th0, the SoC 30 may execute the same process as when the absolute value of the phase difference is smaller than the threshold value Th0, or may execute the same process as when the absolute value of the phase difference is larger than the threshold value Th0. Hereinafter, when the absolute value of the phase difference is equal to the threshold value Th0, the SoC 30 executes the same process as when the absolute value of the phase difference is smaller than the threshold value Th0.

[0102] Furthermore, when the absolute value of the phase difference is greater than the threshold value Th0, if the amplitude square ratio Rn is equal to 0.5, the SoC 30 may execute the same process as when the amplitude square ratio Rn is greater than 0.5, or may execute the same process as when the amplitude square ratio Rn is smaller than 0.5. Hereinafter, if the amplitude square ratio Rn is equal to 0.5, the SoC 30 executes the same process as when the amplitude square ratio Rn is smaller than 0.5.

[0103] 8 is a flowchart showing an example of the positioning control operation according to the first embodiment, which shows a series of operations when the magnetic head 22 passes over one short servo sector SSV.

[0104] First, the SoC 30 obtains a cylinder address from the Gray code GC read by the magnetic head 22 (S101). Next, the SoC 30 performs DFT processing on the read waveform of the N burst NB and the read waveform of the Q burst QB that are read during the period in which the burst gate signal BG is maintained at the "H" level (S102). As a result, the SoC 30 obtains the amplitudes Namp, Qamp, and the phases Nphs, Qphs.

[0105] The SoC 30 calculates the amplitude square ratio Rn based on the amplitudes Namp and Qamp (S103). The SoC 30 calculates the amplitude square ratio Rn using the above-mentioned formula (1).

[0106] The SoC30 executes a folding operation to correct the phases Nphs and Qphs so that the phase difference (Nphs-Qphs) is within 90 degrees (S104). The phase Nphs corrected by the folding operation is denoted as phase Nphs1, and the phase Qphs corrected by the folding operation is denoted as phase Qphs1. In the folding operation, for example, the following procedure is executed. Note that RefPhs is an expected value of the initial phase correction value.

[0107] Nphs1=Nphs,Qphs1=Qphs if Nphs1>=90deg Nphs1=Nphs1-180deg elseif Nphs1<-90deg Nphs1=Nphs1+180deg if Qphs1>=90deg Qphs1=Qphs1-180deg elseif Qphs1<-90deg Qphs1=Qphs1+180deg if abs(Nphs1-Qphs1)>90deg if abs(Qphs1-RefPhs) <abs(Nphs1-RefPhs) if (Nphs1-Qphs1)<0 Nphs1 = Nphs1 + 180 deg else Nphs1=Nphs1-180deg else if (Qphs1-Nphs1)<0 Qphs1=Qphs1+180deg else Qphs1=Qphs1-180deg

[0108] Next, the SoC 30 judges whether the absolute value of the phase difference (Nphs1-Qphs1) is greater than the threshold value Th0 (S105). This process corresponds to the process of judging whether the magnetic head 22 has passed over either the pattern boundary of the N burst NB or the pattern boundary of the Q burst QB.

[0109] If the absolute value of the phase difference (Nphs1-Qphs1) is greater than the threshold value Th0 (S105: Yes), the SoC 30 determines whether the amplitude square ratio Rn is greater than 0.5 (S106). This process corresponds to the process of determining whether the magnetic head 22 has passed over the pattern boundary of the N burst NB or the pattern boundary of the Q burst QB.

[0110] If the amplitude square ratio Rn is greater than 0.5 (S106: Yes), it is estimated that the magnetic head 22 has passed over the pattern boundary of the N-burst NB, and the SoC 30 assigns 1 to the amplitude square ratio Rn (S107).

[0111] If the amplitude squared ratio Rn is not greater than 0.5 (S106: No), it is estimated that the magnetic head 22 has passed over the pattern boundary of the Q burst QB, and the SoC 30 assigns 0 to the amplitude squared ratio Rn (S108).

[0112] For convenience, in the description of FIG. 8, it is assumed that the weight table Wtable is used to calculate the weights Nwt and Qwt regardless of whether the absolute value of the phase difference (Nphs1-Qphs1) is greater than the threshold value Th0. However, when the absolute value of the phase difference (Nphs1-Qphs1) is greater than the threshold value Th0, as described above, SoC30 sets the weights Nwt and Qwt to 0 or 1. According to the relationship defined by the weight table Wtable, when the amplitude square ratio Rn is 0, the weight Nwt can be obtained to be 0 and the weight Qwt to be 1. When the amplitude square ratio Rn is 1, the weight Nwt can be obtained to be 1 and the weight Qwt to be 0. In order to obtain the weights Nwt and Qwt to be 0 or 1 from the weight table Wtable, SoC30 sets the amplitude square ratio Rn to be 0 or 1 in the process of S107 or S108.

[0113] If the absolute value of the phase difference (Nphs1-Qphs1) is greater than the threshold value Th0, the SoC 30 may set the weights Nwt and Qwt to 0 or 1 without using the weighting table Wtable.

[0114] If the absolute value of the phase difference (Nphs1-Qphs1) is not greater than the threshold value Th0 (S105: No), or after the process of S107, or after the process of S108, SoC30 calculates the weights Nwt and Qwt using the weight table Wtable (S109). Specifically, SoC30 obtains the weights Nwt and Qwt using the following formulas (4) and (5). Nwt = Wtable[Rn] (4) Qwt = 1 - Nwt (5)

[0115] Note that Wtable[Rn] included in equation (4) represents a process of acquiring the weight Nwt corresponding to the amplitude squared ratio Rn from the weight table Wtable.

[0116] Next, the SoC 30 calculates an initial phase correction value (S110). The SoC 30 calculates the initial phase correction value RTphs using the following equation (6). RTphs=Nwt*Nphs1+Qwt*Qphs1 ···(6)

[0117] In this specification, "*" is an operator meaning multiplication.

[0118] Next, the SoC 30 executes an initial phase correction operation (S111). The SoC 30 corrects the initial phase using the following equations (7) to (9). The equation (7) is a process for obtaining an initial phase correction value when performing calculations using the sine functions shown in the equations (8) and (9). RTphs2=90deg-RTphs (7) Nbst=Namp*sin(2*pi / 360*(Nphs+RTphs2)) (8) Qbst=Qamp*sin(2*pi / 360*(Qphs+RTphs2)) (9)

[0119] Through this initial phase correction operation, SoC30 can obtain burst data Nbst in which the initial phase shift has been appropriately corrected from the read waveform of N burst NB, and burst data Qbst in which the initial phase shift has been appropriately corrected from the read waveform of Q burst QB.

[0120] The SoC 30 calculates a burst offset based on the burst data Nbst and Qbst (S112).

[0121] The SoC 30 calculates the demodulation position based on the cylinder address obtained by the process of S101 and the burst offset obtained by the process of S112 (S113). Furthermore, the SoC 30 corrects the demodulation position using the postcode PC (S114).

[0122] The SoC 30 executes positioning of the magnetic head 22 (S115) by using the corrected demodulated position as an estimated current position of the magnetic head 22. Then, the operation when the magnetic head 22 passes over one short servo sector SSV is completed.

[0123] Fig. 9 is a diagram showing the initial phase obtained by the series of operations shown in Fig. 8. In this figure, the horizontal axis indicates the offset amount from the track center of the servo track 41 specified by the Gray code GC, and the vertical axis indicates the initial phase obtained by the series of operations shown in Fig. 8. Note that the example shown in this figure corresponds to the examples shown in each part of Fig. 5.

[0124] As shown in Fig. 9, an initial phase of approximately minus 40 degrees is obtained regardless of the offset amount. Even at position L1, which is the pattern boundary of the Q burst QB and where the amplitude Qamp is significantly different from 0, a value close to approximately minus 40 degrees is obtained as the initial phase. In other words, it can be seen that the initial phase is obtained with high accuracy.

[0125] In this way, since the initial phase can be determined with high accuracy even at the pattern boundary, deterioration of the accuracy of positioning control at the pattern boundary is suppressed and deterioration of the accuracy of positioning control is improved.

[0126] As described above, according to the first embodiment, when the magnetic head 22 passes over the short servo sector SSV, the SoC30 identifies the read timing of the N burst NB and the Q burst QB, and reads the N burst NB and the Q burst QB from the short servo sector SSV at the identified read timing. The SoC30 obtains the amplitude Namp, the phase Nphs, the amplitude Qamp, and the phase Qphs by demodulating the read waveform of the N burst NB and the read waveform of the Q burst QB. The SoC30 obtains the weights Nwt and Qwt based on the amplitude Namp, the phase Nphs, the amplitude Qamp, and the phase Qphs. Then, the SoC30 corrects the initial phases of the read waveform of the N burst NB and the read waveform of the Q burst QB by calculating the weighted average using the phases Nphs and Qphs, and the weights Nwt and Qwt. The SoC 30 performs positioning of the magnetic head 22 based on the read waveform of the N burst NB and the read waveform of the Q burst QB after the correction of the initial phase.

[0127] Therefore, even if the amplitude Namp or the amplitude Qamp deviates significantly from 0 at the pattern boundary of the N burst NB or the pattern boundary of the Q burst QB, it is possible to control the weights Nwt and Qwt to 1 or 0 based on the phases Nphs and Qphs. It is possible to accurately calculate the initial phase at each pattern boundary, improving the positioning accuracy at each pattern boundary. In other words, it is possible to obtain a magnetic disk device with high accuracy of positioning control.

[0128] Also, according to the first embodiment, SoC30 determines whether magnetic head 22 has passed over the pattern boundary of N burst NB or the pattern boundary of Q burst QB based on phase Nphs (more precisely, phase Nphs1) and phase Qphs (more precisely, phase Qphs1). If it is determined that magnetic head 22 has passed over the pattern boundary of N burst NB, SoC30 sets weight Nwt to 0. If it is determined that magnetic head 22 has passed over the pattern boundary of Q burst QB, SoC30 sets weight Qwt to 0.

[0129] Therefore, the influence of a phase that is significantly different from the initial phase on the calculated value of the initial phase at the pattern boundary can be completely suppressed, and as a result, the positioning accuracy at each pattern boundary is improved.

[0130] Also, according to the first embodiment, if the absolute value of the difference between phase Nphs (more precisely, phase Nphs1) and phase Qphs (more precisely, phase Qphs1) is greater than threshold value Th0, SoC30 determines that magnetic head 22 has passed over the pattern boundary of N burst NB or the pattern boundary of Q burst QB. If the absolute value of the difference between phase Nphs (more precisely, phase Nphs1) and phase Qphs (more precisely, phase Qphs1) is less than threshold value Th0, SoC30 determines that magnetic head 22 has not passed over either the pattern boundary of N burst NB or the pattern boundary of Q burst QB.

[0131] Furthermore, according to the first embodiment, when the absolute value of the difference between the phase Nphs (more precisely, the phase Nphs1) and the phase Qphs (more precisely, the phase Qphs1) is greater than the threshold value Th0, the SoC30 operates as follows. That is, when the amplitude square ratio Rn is smaller than 0.5, in other words, when the value obtained by squaring the amplitude Namp is smaller than the value obtained by squaring the amplitude Qamp, the SoC30 determines that the magnetic head 22 has passed over the pattern boundary of the N burst NB. When the amplitude square ratio Rn is greater than 0.5, in other words, when the value obtained by squaring the amplitude Namp is greater than the value obtained by squaring the amplitude Qamp, the SoC30 determines that the magnetic head 22 has passed over the pattern boundary of the Q burst QB.

[0132] Furthermore, according to the first embodiment, when the absolute value of the difference between the phase Nphs (more precisely, the phase Nphs1) and the phase Qphs (more precisely, the phase Qphs1) is smaller than the threshold value Th0, the SoC30 operates as follows: That is, the SoC30 acquires the weights Nwt and Qwt based on the weight table Wtable that defines the relationship between the amplitude square ratio Rn and the weights Nwt and Qwt such that the closer the amplitude square ratio Rn is to 1, the closer the weight Nwt is to 1 and the weight Qwt is to 0, and the closer the amplitude square ratio Rn is to 0, the closer the weight Nwt is to 0 and the weight Qwt is to 1.

[0133] In the first embodiment, when it is determined that the magnetic head 22 has passed over a location that is not a pattern boundary, the SoC 30 acquires the weights Nwt and Qwt based on the amplitude square ratio Rn and the weight table Wtable. The method of acquiring the weights Nwt and Qwt when it is determined that the magnetic head 22 has passed over a location that is not a pattern boundary is not limited to this.

[0134] For example, the weight table (referred to as weight table Wtable') is configured to associate the ratio of the amplitude Namp to the amplitude Qamp with the weight Nwt or the weight Qwt instead of the amplitude square ratio Rn. In such a case, the SoC 30 obtains the weights Nwt and Qwt based on the ratio of the amplitude Namp to the amplitude Qamp and the weight table Wtable'.

[0135] In this manner, the SoC 30 is configured to obtain the weights Nwt, Qwt based on the amplitudes Namp, Qamp when it is determined that the magnetic head 22 has passed over a location that is not a pattern boundary.

[0136] According to the first embodiment, the multiple servo sectors SV provided in the circumferential direction of the magnetic disk 11 include normal servo sectors NSV and short servo sectors SSV. The SoC 30 identifies the read timing of the N burst NB and the Q burst QB in the short servo sector SSV based on the timing at which the servo mark SM is detected in the normal servo sector NSV. Then, the SoC 30 executes the series of operations shown in FIG. 8 based on the read waveform of each burst pattern read at the read timing.

[0137] 8 when the magnetic head 22 passes over the normal servo sector NSV. When the magnetic head 22 passes over the normal servo sector NSV, the SoC 30 may skip the processes of S103 to S111 and perform the processes of S112 and after using the burst data before the initial phase correction operation.

[0138] The first embodiment can also be applied to a magnetic disk drive in which all servo sectors SV are configured as normal servo sectors NSV. In such a magnetic disk drive, the SoC may execute the series of operations shown in FIG. 8 for each servo sector SV.

[0139] In addition, in a magnetic disk device in which all servo sectors SV are configured as normal servo sectors NSV, the SoC may determine the timing of rising the burst gate signal BG in a certain servo sector SV based on the servo mark found signal SMF in another servo sector SV. When the SoC determines the timing of rising the burst gate signal BG in another servo sector SV based on the servo mark found signal SMF in a certain servo sector SV, the SoC may execute positioning control by the series of operations shown in FIG.

[0140] Second Embodiment In the first embodiment, one weight table Wtable is set in the SoC 30, and the SoC 30 obtains the weights Nwt and Qwt using the one weight table Wtable. The SoC 30 may be configured to be able to use multiple weight tables.

[0141] In the second embodiment, an example in which the SoC is configured to use multiple weight tables will be described. Note that, here, differences from the first embodiment will be described, and descriptions of the same matters as the first embodiment will be omitted or will be briefly described. The SoC according to the second embodiment will be denoted as SoC30a.

[0142] 10 is a diagram for explaining an example of a method for selecting a plurality of weighting tables according to the second embodiment, in which the horizontal axis indicates the offset amount from the track center, and the vertical axis indicates the absolute value of the phase difference (Nphs1-Qphs1).

[0143] 10, in the second embodiment, a plurality of thresholds Th0 to Th3 including a threshold Th0 corresponding to the first threshold according to the first embodiment are set in the SoC 30a. Threshold Th1 is smaller than threshold Th0, threshold Th2 is smaller than threshold Th1, and threshold Th3 is smaller than threshold Th2.

[0144] If the absolute value of the phase difference is greater than threshold value Th0, SoC 30a uses weighting table Wtable1. If the absolute value of the phase difference is less than threshold value Th0 and greater than threshold value Th1, SoC 30a uses weighting table Wtable2. If the absolute value of the phase difference is less than threshold value Th1 and greater than threshold value Th2, SoC 30a uses weighting table Wtable3. If the absolute value of the phase difference is less than threshold value Th2 and greater than threshold value Th3, SoC 30a uses weighting table Wtable4. If the absolute value of the phase difference is less than threshold value Th3, SoC 30a uses weighting table Wtable5.

[0145] When the absolute value of the phase difference is equal to the threshold value Th0, the SoC 30a may use weighting table Wtable1 or may use weighting table Wtable2. When the absolute value of the phase difference is equal to the threshold value Th1, the SoC 30a may use weighting table Wtable2 or may use weighting table Wtable3. When the absolute value of the phase difference is equal to the threshold value Th2, the SoC 30a may use weighting table Wtable3 or may use weighting table Wtable4. When the absolute value of the phase difference is equal to the threshold value Th3, the SoC 30a may use weighting table Wtable4 or may use weighting table Wtable5.

[0146] In this way, in the second embodiment, the SoC 30a selects a weighting table to be used from among the weighting tables Wtable1 to Wtable5 based on the absolute value of the phase difference. In the following description of the weighting table WtableX (X is an integer between 1 and 5 in the example of FIG. 10), X is referred to as a table number.

[0147] FIG. 11 is a diagram showing an example of the configuration of a plurality of weighting tables Wtable1 to Wtable5 according to the second embodiment.

[0148] Weight table Wtable1, which is a weight table used when the absolute value of the phase difference is greater than threshold value Th0, specifies the relationship between the amplitude square ratio Rn and the weight Nwt such that the weight Nwt is set to 0 in a first range 300 close to 0 in the range of the amplitude square ratio Rn from 0 to 1, and the weight Nwt is set to 1 in a second range 301 close to 1 in the range of the amplitude square ratio Rn from 0 to 1. Since the weight Nwt and the weight Qwt satisfy the relationship shown in equation (5), according to weight table Wtable1, the weight Qwt is set to 1 in the first range 300, and the weight Qwt is set to 0 in the second range 301.

[0149] That is, the weighting table Wtable1 is configured so that the weights Nwt and Qwt are set to 0 or 1 within a predetermined range of the amplitude-squared ratio Rn close to 0 or 1. Therefore, even if the amplitude-squared ratio Rn at the pattern boundary deviates significantly from an ideal value (i.e., 0 or 1), the SoC30a can completely suppress the influence of a large deviation of the phase of the read waveform at the pattern boundary from the actual initial phase.

[0150] Each of the weighting tables Wtable2 to Wtable5 is configured such that when the amplitude square ratio Rn is 0, the weighting Nwt is set to 0, and when the amplitude square ratio Rn is 1, the weighting Nwt is set to 1, so that the closer the amplitude square ratio Rn is to 0, the closer the weighting Nwt is to 0, and the closer the amplitude square ratio Rn is to 1, the closer the weighting Nwt is to 1. The larger the table number of the weighting table, the more gentle the slope of the weighting Nwt with respect to the amplitude square ratio Rn. In other words, the smaller the table number of the weighting table, the wider the range of the amplitude squared ratio Rn in which the weights Nwt, Qwt are set to 0 or 1 or to values ​​close to 0 or 1. The larger the table number of the weighting table, the narrower the range of the amplitude squared ratio Rn in which the weights Nwt, Qwt are set to 0 or 1 or to values ​​close to 0 or 1.

[0151] The weighting tables Wtable1 to Wtable5 are an example of a plurality of pieces of first weighting relationship information configured to enable selection of one piece of first weighting relationship information based on the absolute value of the phase difference. The weighting table Wtable1 is an example of second weighting relationship information.

[0152] The number of weight tables set in the SoC 30a is arbitrary as long as it is two or more.

[0153] FIG. 12 is a flowchart showing an example of the positioning control operation according to the second embodiment.

[0154] The positioning control operation of the SoC 30a differs from the positioning control operation of the SoC 30 according to the first embodiment shown in FIG. 8 in that the processes of S201 to S202 are executed instead of the processes of S105 to S109.

[0155] In S201, the SoC 30a selects a weighting table based on the absolute value of the phase difference (Nphs1-Qphs1). Specifically, the SoC 30a selects one of the weighting tables Wtable1-Wtable5 based on a comparison between the absolute value of the phase difference (Nphs1-Qphs1) and thresholds Th0-Th3.

[0156] In S202, the SoC 30a calculates the weights Nwt and Qwt using the selected weight table. Specifically, the SoC 30a obtains the weights Nwt and Qwt using the following formulas (10) and (11). Nwt=Wtable[TblNo][Rn] ···(10) Qwt = 1 - Nwt (11)

[0157] TblNo is the table number of the weighting table selected by the process of S201. Wtable[TblNo][Rn] represents the process of acquiring the weighting Nwt corresponding to the magnitude squared ratio Rn from the weighting table with the table number TblNo.

[0158] Thus, in the second embodiment, a plurality of weighting tables Wtable1 to Wtable5 are set in the SoC 30a. The plurality of weighting tables Wtable1 to Wtable5 are configured so that one of them can be selected based on the absolute value of the phase difference.

[0159] Depending on the quality of the pattern, even if the absolute value of the phase difference does not reach the threshold value Th0, if the absolute value of the phase difference is close to the threshold value Th0, there is a possibility that the pattern may be influenced by a phase that deviates from the initial phase.

[0160] In order to deal with this, it is conceivable to reduce the threshold value Th0. The technology in which the threshold value Th0 is reduced is referred to as the second comparative example. According to the second comparative example, by lowering the threshold value Th0, the region in which only one of the phases Nwt and Qwt is used becomes wider. Therefore, the accuracy of the calculation of the initial phase correction value may decrease because both the phases Nwt and Qwt are not used.

[0161] In contrast, according to the second embodiment, the weighting tables Wtable1 to Wtable5 are configured to be able to select one of them based on the absolute value of the phase difference. Therefore, the area in which only one of the phases Nwt and Qwt is used can be minimized.

[0162] Therefore, according to the second embodiment, it is possible to suppress the influence of the phase that is significantly different from the initial phase on the calculated value of the initial phase at the pattern boundary, and to widen the region in which both phases Nwt and Qwt are used. Therefore, it is possible to calculate a more accurate initial phase correction value compared to the second comparative example.

[0163] In the first and second embodiments, the SoC 30, 30a used one or more weighting tables as information defining the relationship between the amplitude square ratio and the weighting. The SoC 30, 30a may use any information instead of the one or more weighting tables, as long as the information defines the relationship between the amplitude square ratio and the weighting. For example, the SoC 30, 30a uses one or more formulas or functions defining the relationship between the amplitude square ratio and the weighting, instead of the one or more weighting tables. The information defining the relationship between the amplitude square ratio and the weighting is an example of weighting relationship information.

[0164] In the first and second embodiments, a common value may be used as the first threshold value Th0 in the entire area of ​​the magnetic disk 11, or different values ​​may be used as the first threshold value Th0 depending on the position on the magnetic disk 11. For example, the recording surface of the magnetic disk 11 is divided into a plurality of radial areas in the radial direction. Then, the value of the first threshold value Th0 is set for each radial area depending on the writing quality of each unit pattern. The SoCs 30 and 30a may autonomously set the value of the first threshold value Th0, or the value of the first threshold value Th0 may be set by a designer.

[0165] As for the threshold values ​​Th1 to Th3 in the second embodiment, similarly to the first threshold value Th0, a common value may be used for the entire area of ​​the magnetic disk 11, or different values ​​may be used as the first threshold value Th0 depending on the position on the magnetic disk 11. The SoC 30a may autonomously set the values ​​of the threshold values ​​Th1 to Th3, or the threshold values ​​Th1 to Th3 may be set by a designer.

[0166] Although some embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0167] 1 magnetic disk device, 2 host, 11 magnetic disk, 12 SPM, 13 lamp, 15 actuator arm, 21 SVC, 22 magnetic head, 22r read head, 22w write head, 23 HDC, 24 preamplifier, 25 RWC, 26 processor, 28 FROM, 29 DRAM, 30 controller, 41 servo track, 100 range, 200 relationship, 300 first range, 301 second range, SG servo gate signal, BG burst gate signal, SMF servo mark found signal, PA preamble, SM servo mark, GC gray code, NB N burst, QB Q burst, PC post code, SV servo sector, NSV normal servo sector, SSV short servo sector.

Claims

1. a magnetic disk having a plurality of servo sectors provided in a circumferential direction, each of the plurality of servo sectors having a first burst pattern and a second burst pattern recorded therein; a magnetic head for writing and reading data to and from the magnetic disk; When the magnetic head passes over a first servo sector that is one of the plurality of servo sectors, Identifying read timings of the first burst pattern and the second burst pattern; reading the first burst pattern and the second burst pattern from the first servo sector at the read timing; demodulating a first read waveform that is a read waveform of the first burst pattern and a second read waveform that is a read waveform of the second burst pattern to obtain a first amplitude and a first phase of the first read waveform and a second amplitude and a second phase of the second read waveform; obtaining a weight based on the first amplitude, the second amplitude, the first phase, and the second phase; correcting initial phases of the first read waveform and the second read waveform based on an initial phase correction value obtained by calculating a weighted average using the weight, the first phase, and the second phase; performing positioning of the magnetic head based on the first read waveform and the second read waveform after the correction of the initial phase; A controller; A magnetic disk device comprising:

2. the first burst pattern and the second burst pattern each include a plurality of first unit patterns and a plurality of second unit patterns, and have a configuration in which the first unit patterns and the second unit patterns are alternately arranged in a radial direction and a circumferential direction, respectively; a radial position of a first boundary and a radial position of a second boundary are different from each other, the first boundary is a boundary between a first unit pattern and a second unit pattern in a radial direction in the first burst pattern, and the second boundary is a boundary between a first unit pattern and a second unit pattern in a radial direction in the second burst pattern, the weights include a first weight applied to the first phase and a second weight applied to the second phase; The controller: determining whether the magnetic head has passed over the first boundary or the second boundary based on the first phase and the second phase; setting the first weight to 0 when it is determined that the magnetic head has passed over the first boundary, and setting the second weight to 0 when it is determined that the magnetic head has passed over the second boundary.

2. The magnetic disk drive according to claim 1.

3. The controller: obtaining the first weight and the second weight based on the first amplitude and the second amplitude when it is determined that the magnetic head has not passed over either the first boundary or the second boundary; 3. The magnetic disk drive according to claim 2.

4. The controller: if an absolute value of a difference between the first phase and the second phase is greater than a first threshold value, it is determined that the magnetic head has passed over the first boundary or the second boundary; if the absolute value is smaller than the first threshold value, it is determined that the magnetic head has not passed over either the first boundary or the second boundary; 3. The magnetic disk drive according to claim 2.

5. The controller: if the absolute value is greater than the first threshold value, If the value obtained by squaring the first amplitude is smaller than the value obtained by squaring the second amplitude, it is determined that the magnetic head has passed over the first boundary; if the square of the first amplitude is greater than the square of the second amplitude, it is determined that the magnetic head has passed over the second boundary.

5. The magnetic disk drive according to claim 4.

6. The controller: If the absolute value is less than the first threshold value, the first weight and the second weight are obtained using weight relationship information that defines a relationship between the ratio and the weight such that the closer a ratio obtained by dividing the value obtained by squaring the first amplitude by the sum of the value obtained by squaring the first amplitude and the value obtained by squaring the second amplitude is to 1, the closer the first weight is to 1 and the closer the second weight is to 0, and the closer the ratio is to 0, the closer the first weight is to 0 and the closer the second weight is to 1; 5. The magnetic disk drive according to claim 4.

7. The controller: weight relationship information is set that defines a relationship between a ratio obtained by dividing a value obtained by squaring the first amplitude by a sum of a value obtained by squaring the first amplitude and a value obtained by squaring the second amplitude, and the first weight and the second weight; obtaining the first weight and the second weight based on weight relationship information based on an absolute value of a difference between the first phase and the second phase, and based on the first amplitude and the second amplitude; 3. The magnetic disk drive according to claim 2.

8. The weight relationship information is a plurality of pieces of first weight relationship information defining a relationship between the ratio and the first weight and the second weight, and configured to be able to select one of the plurality of pieces of first weight relationship information based on an absolute value of the phase difference; 8. The magnetic disk drive according to claim 7.

9. The plurality of first weight relationship information and second weight relationship information in which a relationship between the ratio and the weights is defined such that, when the ratio is within a first range close to 0 in the range from 0 to 1, the first weight is set to 0 and the second weight is set to 1, and, when the ratio is within a second range close to 1 in the range from 0 to 1, the first weight is set to 1 and the second weight is set to 0.

9. The magnetic disk drive according to claim 8.

10. the plurality of servo sectors include a second servo sector and a third servo sector; the second servo sector has a servo mark recorded therein in addition to the first burst pattern and the second burst pattern; the first servo sector corresponds to the third servo sector, the controller specifies the read timing based on a timing at which the servo mark is detected in the second servo sector; The magnetic disk drive according to any one of claims 1 to 9.

11. No servo mark is recorded in the third servo sector. The magnetic disk drive according to claim 10.

Images