JP4504338B2 - Control method, control circuit, and storage device - Google Patents

Description

  The present invention relates to a signal reproduction method and a storage device, and more particularly to a signal reproduction method for improving a phase margin of a control system and a storage device employing such a signal reproduction method. The present invention also relates to a head control method and a control circuit.

  In the case of a storage device having a RAID (Redundant Array of Independent Disks) configuration, a plurality of hard disk drives (HDDs) are mounted. In such a storage device, it is often the case that a write is executed in an on-track state in one HDD and a seek to a target cylinder is executed in another HDD at the same timing. In this case, the HDD performing the seek is shaken by the vibration of the HDD performing the seek to be in an off-track state, resulting in a problem that the performance of the storage device is degraded. This problem is called “box vibration problem” or “rocker vibration problem”.

In recent storage devices, as the recording density has increased due to the increase in track per inch (TPI), it has become easy to be in an off-track state even with a small disturbance, and the head has been increased with an increase in access time. It has become essential to suppress disturbances by increasing the bandwidth of the control system due to the fact that a large current flows during acceleration and deceleration and the excitation force has increased.
JP-A-5-258492 JP-A-7-161158 International Publication No. 99/36907 Pamphlet JP 60-10472 A JP 2000-123506 A

  However, in the reproducing system of the storage device, the arithmetic processing for generating a signal for controlling the position of the head necessarily involves an arithmetic delay. For this reason, the phase margin of the control system is wasted due to this calculation delay, and there is a first problem that the gain of the control system cannot be increased easily.

  Further, when the gain of the control system is simply increased, there is a second problem that the phase margin is lost and the closed loop oscillates.

  The prior art has a third problem that it is difficult to detect the speed quickly and accurately especially when the speed of the magnetic head is increased.

  Therefore, the present invention provides a control method, a control circuit, a signal reproduction method, and a storage device that secure the phase margin of the control system and solve the first problem, the second problem, and / or the third problem. The purpose is to do.

  The above-described problem is a signal reproduction method for reproducing servo information recorded on a recording medium using a head, and relates to a current calculation regarding current from the reproduced servo information and a position and speed of the head on the recording medium. A matrix calculation step for performing an observer calculation, and a filter calculation step for calculating a current value for driving the head based on a calculation result of the matrix calculation step. The filter calculation step demodulates the position of the head. During the first period from when the current value is output, only the calculation that requires the demodulation result of the current sample is performed, and the calculation that requires the demodulation result of the past sample is performed in other than the first period. This is achieved by a signal reproduction method characterized in that it is performed during the second period. According to the signal reproduction method of the present invention, the first problem can be solved.

  In this case, the filter calculation step may perform a vector shift process that requires a past sample during the second period.

  The above problem is a signal reproduction method for reproducing servo information recorded on a recording medium using a head, and estimating the head from the actual position of the head on the recording medium obtained from the reproduced servo information. This can be achieved by a signal reproduction method comprising the steps of obtaining a prediction error by subtracting the position and feeding back the prediction error to a control system for calculating a current value for driving the head. According to the signal reproduction method of the present invention, the second problem can be solved.

  In this case, the signal reproduction method may further include the step of concentrating the feedback of the prediction error on the first half of the current multirate output during multirate control in which the current value is output in a plurality of times.

  The above-described problem is a signal reproduction method for reproducing servo information recorded on a recording medium using a head, and phase information included in the servo information for demodulating the actual position of the head on the recording medium. Can be achieved by a signal reproduction method characterized by including a position demodulating step for demodulating the head and a speed obtaining step for obtaining the speed of the head based on the phase difference of the demodulated phase information. According to the signal reproduction method of the present invention, the third problem can be solved.

  The above-described problems include reproducing means for reproducing servo information recorded on a recording medium using a head, current calculation relating to current from the reproduced servo information, and observer calculation relating to the position and velocity of the head on the recording medium. Matrix calculation means for performing the calculation, and filter calculation means for calculating the current value for driving the head based on the calculation result of the matrix calculation means, the filter calculation means after demodulating the position of the head During the first period until the current value is output, only the calculation that requires the demodulation result of the current sample is performed, and the calculation that requires the demodulation result of the past sample is performed in the second period other than the first period. This can be achieved by a storage device characterized in that it is performed during According to the storage device of the present invention, the first problem can be solved.

  In this case, the filter calculation means may be configured to perform vector shift processing that requires past samples during the second period.

  The above-described problems include reproducing means for reproducing servo information recorded on a recording medium using a head, and subtracting the estimated position of the head from the actual position of the head on the recording medium obtained from the reproduced servo information. This can be achieved by a storage device comprising a means for obtaining a prediction error and a feedback means for feeding back the prediction error to a control system for calculating a current value for driving the head. According to the storage device of the present invention, the second problem can be solved.

  In this case, the storage device may further include a control unit that concentrates feedback of the prediction error on the first half of the current multirate output during multirate control in which the current value is output in a plurality of times. .

  The above-described problems include a reproducing means for reproducing servo information recorded on a recording medium using a head, and demodulating phase information included in the servo information in order to demodulate the actual position of the head on the recording medium. This can be achieved by a storage device comprising position demodulating means and speed acquisition means for obtaining the speed of the head based on the phase difference of the demodulated phase information. According to the storage device of the present invention, the third problem can be solved.

  The above problem is a head control method for reproducing servo information recorded on a recording medium, and subtracting the estimated position of the head from the actual position of the head on the recording medium obtained from the reproduced servo information A prediction error, a first feedback step of adding the input prediction value and feeding back to the control system for calculating a current value for driving the head, and a predetermined multiplication coefficient for the prediction error A second feedback step that feeds back to the control system only through a multiplier that multiplies the first feedback step, an addition step that adds the first feedback step, the output value, and the output value of the second feedback step; It can achieve by the control method characterized by including.

  The above-mentioned problems include a means for subtracting the estimated position of the head from the actual position of the input head to obtain a prediction error, and a control system for calculating a current value for driving the head by adding the input predicted value First feedback means for feeding back to the control system, second feedback means for feeding back to the control system only through a multiplier for multiplying the prediction error by a predetermined multiplication coefficient, the first feedback means and the output value This can be achieved by a control circuit comprising addition means for adding the output value of the second feedback means.

  The above-described problems include reproducing means for reproducing servo information recorded on a recording medium using a head, and subtracting the estimated position of the head from the actual position of the head on the recording medium obtained from the reproduced servo information. A means for obtaining a prediction error, a first feedback means for adding back the inputted prediction value to a control system for calculating a current value for driving the head, and multiplying the prediction error by a predetermined multiplication coefficient Second feedback means that feeds back to the control system only through a multiplier that performs the operation, and addition means for adding the first feedback means, the output value, and the output value of the second feedback means. It can be achieved by a storage device characterized by

  Therefore, according to the present invention, it is possible to ensure the phase margin of the control system, it is possible to suppress the operation delay or increase the gain of the control system, and even if the speed of the magnetic head is increased, It is also possible to detect the speed quickly and accurately.

  Embodiments of a control method, a control circuit, a signal reproduction method, and a storage device according to the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram showing a first embodiment of a storage device according to the present invention. The first embodiment of the storage device employs the first embodiment of the signal reproducing method according to the present invention. In this embodiment, the present invention is applied to a magnetic disk device having a RAID configuration.

  In FIG. 1, the magnetic disk apparatus is roughly composed of a micro controller unit MCU1, a digital signal processor (DSP) 2, a user logic circuit 3, a voice coil motor (VCM) control circuit 4, and a spindle motor (SPM) control circuit 5. , A read / write channel (RDC) 6, a servo demodulation circuit 7, and a disk enclosure (DE) 8. The disk enclosure 8 includes a VCM 11, an SPM 12, a head IC 13, an arm 14 on which a magnetic head 16 is mounted, and a magnetic disk 15. For convenience of explanation, only one magnetic disk 15 is shown in FIG. 1, but since a plurality of magnetic disks 15 are actually provided, a plurality of VCMs 11, arms 14 and magnetic heads 16 are also provided in accordance with the number of magnetic disks 15. Is provided.

  The head IC 13, the read / write channel 6, the servo demodulation circuit 7, the DSP 2, and the user logic circuit 3 constitute a reproduction system. The control system has a VCM control circuit 4 and an SPM control circuit 5 in addition to the reproduction system. The recording system includes the DSP 2, the user logic circuit 3, the read / write channel 6, and the head IC 13. However, since the recording system is not directly related to the present invention, the description thereof is omitted. Further, the DSP 2, the user logic circuit 3 and the VCM control circuit 4 constitute circuit means for instructing a current value to be supplied to the VCM 11. The head IC 13, the read / write channel 6 and the servo demodulation circuit 7 are on the magnetic disk 15. The circuit means for demodulating the current position (position) is configured.

  The DSP 2 receives, for example, a seek command from the MCU 1 via the user logic circuit 3. The DSP 2 calculates a current value for moving the arm 14 to the cylinder position and the head position specified by the seek command, and gives a current instruction to the VCM control circuit 4. The VCM control circuit 4 generates a current for driving the VCM 11, and the arm 14 moves according to the current value. After the movement operation to the designated cylinder position and head position is completed, the on-track operation is continued at that position until the next command is supplied from the MCU 1.

  The SPM control circuit 5 generates a control signal for rotating the magnetic disk 15 at a specified rotation speed from the DSP 2 via the user logic circuit 3 and supplies the control signal to the SPM 12. Thereby, the magnetic disk 15 is rotated at the rotation speed designated by the SPM 12.

  The head IC 13 serves as an amplifier that amplifies the read signal read from the magnetic disk 15 by the magnetic head 16. The read / write channel 6 performs read signal filtering, automatic gain control (AGC) processing, etc., and shapes the signal waveform. The servo demodulation circuit 7 demodulates the current position (position) on the magnetic disk 15 in accordance with the demodulation algorithm for the signal waveform shaped by the read channel 6. The DSP 2 determines the current value to be output next based on the demodulation result of the servo demodulation circuit 7. VCM control is performed by such a closed loop.

  FIG. 2 is a diagram for explaining servo information on the magnetic disk 15. As shown in the figure, since this embodiment employs an embedded servo system, servo information is stored in the servo area 15-1 in the radial direction of the magnetic disk 15, as shown in the figure. Data is recorded so as to partition a data recording area 15-2 in which data is recorded. As described above, since the servo information is intermittently recorded during one round of the magnetic disk 15, a servo gate signal is generated as a window for demodulating the servo information from the reproduction signal. Generation and control of the servo gate signal is performed by a circuit in the user logic circuit 3. A voltage controlled oscillator (VCO) counter that counts the time from one servo information to the next servo information is also provided in the user logic circuit 3. The DSP 2 resets the VCO counter, sets the maximum count value, and controls the on / off timing of the servo gate signal.

  FIG. 3 is a timing chart for explaining the operation delay. In the figure, (A) shows the count value of the VCO counter, and (B) shows the servo gate signal. TA represents a period from the start of interrupt to position demodulation, TB represents a period from position demodulation to current value output, and TC represents a period from current value output to next interrupt generation. The arithmetic delay corresponds to the period TB. If the arithmetic delay becomes large, the phase margin of the control system is wasted, so the arithmetic delay needs to be made as small as possible. That is, it is important to minimize the arithmetic processing performed within the period TB.

  FIG. 4 is a flowchart for explaining a main part of processing performed by the reproduction system. This figure shows the processing of the DSP 2 performed in the period TB in FIG.

  In FIG. 4, when an interrupt is started in step S1, step S2 performs matrix calculation. Matrix calculation includes current calculation for calculating the current in the magnetic disk device and observer calculation for determining the current position (position) on the magnetic disk 15, the current speed of the magnetic head 16, and the like. In step S3, filter calculation is performed based on the matrix calculation result, and a current value instructed to the VCM control circuit 4 to drive the VCM 11 is calculated. In step S4, the calculated current value is instructed to the user logic circuit 3. Note that only the current calculation of the matrix calculation is performed in the period TB and the observer calculation of the matrix calculation is performed in the period TC, so that the processing time can be shortened to some extent and the operation delay can be reduced.

  In this embodiment, a part of the filter calculation is performed in the period TA and / or the period TC, and only the remaining part of the filter calculation is performed in the period TB, thereby further reducing the processing time of the calculation performed in the period TB.

The n-th order filter calculation can be expressed by the following equation (1) where u (k) is an input and y (k) is an output. In equation (1), a _i represents a filter denominator coefficient, and b _i represents a filter numerator coefficient.

式（１）

式（１）は、次式（２）の形で表すことができる。

Formula (1)

Formula (1) can be expressed in the form of the following formula (2).

式（２）

従って、過去のサンプルに基いて行える次式（３）の計算を期間Ａ及び/期間Ｃで行い、ポジションを復調した後の期間Ｂでは現在のサンプルの復調結果を必要とする次式（４）の計算を行うことで、期間Ｂにおいて行われる演算の処理時間を大幅に短縮することができ、この処理時間の短縮効果は、フィルタの次数ｎが大きくなる程大きい。

Formula (2)

Therefore, the calculation of the following equation (3) that can be performed based on the past samples is performed in the periods A and / or C, and the following equation (4) that requires the demodulation result of the current sample in the period B after demodulating the position. By performing the calculation, it is possible to significantly shorten the processing time of the calculation performed in the period B, and the effect of shortening the processing time becomes larger as the order n of the filter increases.

式（３）

Formula (3)

式（４）

更に、図５に示すように、過去のサンプルのシフト処理も、ＤＳＰ２に比較的大きな負荷を与えるので、このシフト処理も期間Ａ及び/又は期間Ｃで行うことにより、期間Ｂにおいて行われる演算の処理時間を更に短縮することができる。

Formula (4)

Further, as shown in FIG. 5, since the past sample shift processing also places a relatively large load on the DSP 2, if this shift processing is also performed in the period A and / or the period C, the calculation performed in the period B can be performed. The processing time can be further shortened.

FIG. 6 is a flowchart for explaining the filter calculation. In the figure, step S11 sets y (k) = b ₀ u (k) + tmp (k). Step S11 constitutes an arithmetic processing portion S20 that performs the first main calculation. In step S12, i = 1 is set. Step S13 determines whether i is n or less, and if the determination result is YES, step S14 is tmp (k + 1) = − a _i y (k + 1−i) + b _i u. (K + 1-i) is calculated, and the process returns to step S13. If the determination result of step S13 is NO, the process proceeds to step S16. Steps S12 to S14 constitute an arithmetic processing portion S21 that performs the second main calculation.

  Step S16 sets i = 1, and step S17 determines whether i is n or less. If the decision result in the step S17 is YES, a step S18 calculates y (ki) = y (ki + 1) and u (ki) = u (ki + 1), The process returns to step S17. The process ends when the decision result in the step S17 becomes NO. Steps S16 to S18 constitute a vector shift processing portion S22 that performs the shift processing.

  Step S11 of the arithmetic processing portion S20 is executed in the period TB. Steps S12 to S14 of the arithmetic processing portion S21 are executed in the period TA and / or the period TC. Further, steps S16 to S18 of the vector shift processing part S22 are executed in the period TA and / or the period TC.

  According to the present embodiment, in the period B from the position demodulation to the output of the current value, about 10 to 20% of the arithmetic processing that requires the demodulation result of the current sample in the filter calculation Therefore, the operation delay can be greatly reduced. In addition, the effect of reducing the operation delay increases as the filter order n increases, and the operation delay can be effectively reduced even if the control unit and filter unit of the storage device become more complex in the future.

  In the filter calculation, the remaining calculation processing period that can be performed based on the past sample may be performed within the period A and / or the period C according to the time margin.

  Next, a second embodiment of the storage device according to the present invention will be described. The second embodiment of the storage device employs a second embodiment of the signal reproducing method according to the present invention. In this embodiment, the present invention is applied to a magnetic disk device having a RAID configuration. The block configuration of the second embodiment of the storage device is the same as the block configuration of the first embodiment shown in FIG.

  FIG. 7 is a functional block diagram for explaining the operation of the main part of the second embodiment. In the figure, the controller 21 comprises a subtractor 31 to which a target position is input, multipliers 32, 33, 34, 35 for multiplying corresponding multiplication coefficients Kp, Kv, Kb, Ke, and adders 36, 37. . The observer 22 includes a VCM model 41, a subtracter 42, multipliers 43, 44, 45 for multiplying corresponding multiplication coefficients Lb, Lv, Lp, and adders 46, 47. The controller 21 and the observer 22 correspond to the DSP 2 in FIG.

  The output u (n) of the controller 21 output from the adder 37 corresponds to the output obtained via the user logic circuit 3 and the VCM control circuit 4 in FIG. Supplied to the VCM model 41 of the observer 22. An acceleration disturbance is also supplied to the adder 24, and the output of the adder 24 is supplied to the control object 25. The control target 25 corresponds to the VCM 11 in FIG.

  The actual position calculation unit 201 corresponds to the magnetic head 16, the head IC 13, the read / write channel 6, and the servo demodulation circuit 7 in FIG. 1, and the obtained actual position is supplied to the subtracter 42 of the observer 22. As a result, the estimated position output from the VCM model 41 is subtracted from the actual position by the subtractor 42, and the output of the subtractor 42 is fed back to the adder 37 via the multiplier 35 of the controller 21 as a prediction error. .

  Note that BiasBar (n) input to the adder 46 is a prediction bias, VelBar (n) input to the adder 47 is a prediction speed, BiasHat (n) output from the adder 46 is an estimation bias, and an adder 47 VelHat (n) output from, indicates the estimated speed, and PosHat (n) output from the multiplier 45 indicates the estimated position. Here, when L is an arbitrary coefficient, the estimated value is expressed by correcting the current predicted value using a prediction error (estimated value = predicted value + (L × predictive error)). The predicted value is calculated based on predicted value = estimated value for bias and v = v0 + at of the physical formula for speed. Therefore, VelBar (n + 1) = VelHat (n) + (BiasHat (n) + u (n)) × ts, and BiasBar (n + 1) = BiasHat (n).

  When there is no acceleration disturbance and there is no modeling error, the prediction error output from the subtractor 42 becomes zero. Therefore, in this embodiment, the prediction error is regarded as a kind of disturbance, and a feedback path for feeding back the prediction error is provided as shown by a thick line in FIG. 7, so that the phase margin of the control system is not wasted. Can be raised. In other words, in the case of a control system designed to have a gain equivalent to that of the conventional control system, it is possible to earn a sufficient margin according to this embodiment. Further, since the present embodiment has a configuration in which a feedback path indicated by a thick line in FIG. 7 is added to the conventional control system, it is not necessary to increase the order of the controller 21 and can be easily realized.

  Next, a modified example of the second embodiment will be described with reference to FIGS. The multi-rate control is to control the movement of the magnetic head 16 more smoothly by outputting current to the VCM 11 in a plurality of times for one servo demodulation operation. When such multi-rate control is performed in the prediction error feedback of the second embodiment, it is desirable to further improve the phase margin. Therefore, in this modification, the feedback of the prediction error is concentrated on the first half of the multirate output.

  FIG. 8 is a diagram showing a main part of a modification of the second embodiment. As shown in the figure, in this modification, a switch 38 is provided on the input side of the multiplier 35. The switch 38 is provided, for example, on the controller 21 side. For example, when quadruple multirate control is performed, prediction error feedback is performed only for the first one of the four current values output to the VCM 11, and the remaining three current values are set. On the other hand, on / off of the switch 38 is controlled so that the prediction error feedback is not performed. In this case, the multiplication coefficient Ke of the multiplier 35 is set to a value about four times that in the case where the prediction error feedback is equally performed on the four current values output to the VCM 1. Thereby, when a prediction error occurs, it can be corrected as quickly as possible.

  FIG. 9 is a timing chart for explaining the operation of a modification of the second embodiment. In the figure, (A) shows a servo gate signal, and (B) shows a current instruction value for the VCM control circuit 4.

  FIG. 10 is a functional block diagram showing the main part of a modification of the second embodiment. In the figure, the DSP 2 includes an actual position calculation unit 201, an estimated position calculation unit 202, a current calculation unit 203, a Ke feedback on / off control unit 204, and a multirate control unit 205. A circuit portion 206 including an estimated position calculation unit 202, a current calculation unit 203, and a Ke feedback on / off control unit 204 surrounded by a broken line corresponds to the controller 21 and the observer 22 shown in FIG.

  The actual position calculation unit 201 calculates and outputs the actual position on the magnetic disk 15 as in the case of FIG. For example, in the case of quadruple multirate control, the actual position calculation unit 201 operates once, while the circuit portion 206 operates four times. The estimated position calculation unit 202 calculates and outputs an estimated position on the magnetic disk 15 as in the case of the VCM model 41 of FIG. The current calculation unit 203 calculates and outputs a current instruction value to be output to the VCM control circuit 4 using the actual position and the estimated position. The Ke feedback on / off control unit 204 performs prediction error feedback only for the first one of the four current values output to the VCM control unit 4 in the case of quadruple multirate control, On / off of prediction error feedback is controlled so that prediction error feedback is not performed for the remaining three current values. The multirate control unit 205 outputs the first current instruction value of the four current instruction values from the current calculation unit 203 to the VCM control circuit 4 at the same time as the calculation by the current calculation unit 203 is completed. The second to fourth current instruction values are held, and when the time for outputting the current value comes based on the count value of the VCO counter, the current value is output to the VCM control circuit 4.

  By the way, the process of shifting from the servo cut to the on-track state is called re-zero process. FIG. 11 is a flowchart for explaining the re-zero process. In step S31, the magnetic head 16 is held at the standby position (or retract position) by flowing a constant current to the VCM 11 in a servo cut state where servo information is not demodulated. When the re-zero process is started in the state of step S31, step S32 and step S33 are performed in parallel. Step S32 performs blindfold seek, and step S33 performs servo mark clock confirmation. The blind seek seeks to seek based on the estimated position on the assumption that the estimated position of the magnetic head 16 is correct, and is also called a blind seek. In the servo mark clock confirmation, the servo information is read by locking the servo information on the magnetic disk 15 at the position of the servo mark. After steps S32 and 33, step S34 confirms the current speed of the magnetic head 16.

  In step S35, a discharge process is performed, and control is performed so that the speed of the magnetic head 16 becomes zero. In step S36, the current position of the magnetic head 16 is confirmed using cylinder information (gray code). In step S 37, fine control is performed to finely adjust the current position of the magnetic head 16 to a desired position on the magnetic disk 15.

  In the rezero processing, it is important that the current speed in step S34 can be confirmed quickly and accurately. However, as the recording density of the magnetic disk device increases, it becomes difficult to confirm the current speed quickly and accurately mainly due to the reasons (1) to (3) described below. .

  Reason (1): As the recording density increases, it is necessary to reduce the ratio of the servo area 15-1 on which the servo information on the magnetic disk 15 is recorded as much as possible. Therefore, the cylinder information (full gray code in one servo frame) ) Cannot be recorded. Therefore, the cylinder information is divided and recorded in a plurality of servo frames. However, in order to demodulate the divided cylinder information, it is necessary to determine the index and determine the current servo frame position, so it takes too much time to check the current speed. End up.

  Reason (2): In addition to cylinder information, position information that repeats the same pattern for each predetermined number of cylinders is recorded on the magnetic disk 15. Therefore, the relative position can be demodulated based on the position information. However, for example, when the same pattern is repeatedly recorded every 32 cylinders, for example, 32 cylinders / sample and 64 cylinders / sample appear to be the same speed because the same pattern is repeated every 32 cylinders. For this reason, if the current speed is confirmed based on the position information, an accurate speed cannot be obtained.

  Reason (3): When the magnetic disk device is not in operation, the magnet catching force for holding the magnetic head 16 in the standby position using a magnet tends to become stronger in order to improve the shock resistance performance of the magnetic disk device. . For this reason, even if the driving force of the arm 14 slightly deviates from the target value due to a temperature change or a change over time, the magnetic head 16 is returned to the standby position by the magnet catching force during the blindfold seek, May reach the limit position in the opposite radial direction of the magnetic disk 15, and the blinding seek will not be performed normally. As a result, the blind seek may actually move at a considerable speed even though it is assumed that the magnetic head 16 stops at a desired position on the magnetic disk 15 at the end of the seek. There is.

  Thus, for the reasons (1) to (3) as described above, it is very difficult to confirm the current speed of the magnetic head 16 quickly and accurately after the blindfold seek.

  Therefore, a description will be given of a third embodiment of the storage device according to the present invention which can detect the current speed of the magnetic head 16 quickly and accurately. The third embodiment of the storage device employs a third embodiment of the signal reproducing method according to the present invention. In this embodiment, the present invention is applied to a magnetic disk device having a RAID configuration. The block configuration of the third embodiment of the storage device is the same as the block configuration of the first embodiment shown in FIG.

  FIG. 12 is a diagram for explaining a servo track writer (STW) pattern of servo information used in this embodiment. When switching from the write state to the read state in order to demodulate the servo information, a decrease in which normal read cannot be performed immediately because the read signal has a transient is called a write-to-read transient. In the figure, a cap region GAP is provided to absorb this write-to-read transient. The servo mark SVMK indicates the start position of the servo information and has a special pattern that does not exist in the data area. In the position area P, phase information for position demodulation is recorded, and is composed of EVEN1, ODD1, and EVEN2 layers as will be described later. In the area HSC, a cylinder number, a magnetic head number, a sector number, and the like are recorded.

  FIG. 13 is a diagram showing a recording pattern of phase information for position demodulation in the position area P. In the figure, a horizontal broken line indicates a boundary line between adjacent tracks, an oblique solid line indicates a maximum value (peak) of phase information, and an oblique broken line indicates a minimum value (valley) of phase information. In the EVEN1 and EVEN2 layers, the angle formed by the above-described mountain and valley patterns with respect to the track is the same, and is different from the angle in the ODD1 layer. The original use of the phase information is to demodulate the current position of the magnetic head 16, that is, the position, from the phase difference between the ODD1 layer and the EVEN1 and EVEN2 layers. Even in the case where the magnetic head 16 has speed, a configuration in which the ODD1 layer is sandwiched between the EVEN1 and EVEN2 layers in the position region P so that the phase difference between the ODD layer and the EVEN layer can be calculated with reference to the center of the ODD layer. It has become. The position is obtained by comparing the sum of the vectors of the EVEN1 layer and the EVEN2 layer with the phase of the vector of the ODD1 layer.

  FIG. 14 is a diagram illustrating a state where the phase difference is zero when the velocity of the magnetic head 16 is zero. 4A shows the scanning trajectory of the magnetic head 16 with an arrow, and FIG. 4B shows the sum of the vector O1 in the ODD1 layer and the vectors E1 and E2 in the EVEN1 and EVEN2 layers. In this case, the phase difference between the ODD1 layer and the EVEN1 and EVEN2 layers is zero.

  FIG. 15 is a diagram for explaining a state in which the phase difference is zero when the velocity of the magnetic head 16 is not zero. 4A shows the scanning trajectory of the magnetic head 16 with an arrow, and FIG. 4B shows the sum of the vector O1 in the ODD1 layer and the vectors E1 and E2 in the EVEN1 and EVEN2 layers. In this case, the phase difference between the ODD1 layer and the EVEN1 and EVEN2 layers is zero.

  In this way, the sum of the vector E1 of the EVEN1 layer and the vector E2 of the EVEN2 layer is obtained and compared with the vector O1 of the ODD1 layer, thereby obtaining the phase difference between the ODD1 layer and the EVEN1 and EVEN2 layers. The position can be demodulated. However, in FIG. 15B, the angle formed by the hatched vectors E1 and E2 is proportional to the moving speed of the magnetic head 16. Therefore, in this embodiment, the current speed of the magnetic head 16 is confirmed using the phase difference between the EVEN1 layer and the EVEN2 layer. However, the speed of the magnetic head 16 that can be obtained from the phase difference between the EVEN1 layer and the EVEN2 layer includes the influence of the rotational fluctuation of the spindle motor 12 and therefore has a relatively low accuracy. Therefore, the accuracy of the velocity of the magnetic head 16 that can be obtained from the phase difference between the EVEN1 layer and the EVEN2 layer can be improved by combining with the position demodulation result. That is, for example, in the case of the 32 cylinder group, the speed is obtained using the position demodulation result for speeds of ± 16 cylinders / sample or less, and the speeds higher than that may be obtained from the phase difference between the EVEN1 layer and the EVEN2 layer.

  FIG. 16 is a flowchart for explaining the operation of the third embodiment. The speed detection process shown in the figure is performed by the DSP 2 shown in FIG.

  In FIG. 16, when the speed detection process is started, step S41 sets Pe1 to the phase of the EVEN1 layer, and step S42 sets Pe2 to the phase of the EVEN2 layer. In step S43, when the speed conversion coefficient is indicated by K, the speed VelEE of the magnetic head 16 is obtained from VelEE = K (Pe1-Pe2). In step S44, it is determined whether or not the speed VelEE is sufficiently small. As described above, for example, in the case of the 32 cylinder group, if the speed VelEE is a speed of ± 16 cylinders / sample or less, the determination result in step S44 is YES, and if the speed is greater than ± 16 cylinders / sample, The determination result is NO. If the decision result in the step S44 is YES, a step S45 obtains the speed of the magnetic head 16 from the difference between the demodulated current position and the previous position, and the process ends. On the other hand, if the decision result in the step S44 is NO, a step S46 assumes that the speed of the magnetic head 16 is the speed VelEE, and the process ends. When the speed of the magnetic head 16 is obtained in step S46, unlike in step S45, the speed can be obtained within one sample.

  According to the present embodiment, even when the speed of the magnetic head 16 is relatively high, the correct speed can be obtained quickly, and the success rate of the rezero process can be improved. Further, the speed detection process in the present embodiment is not limited to the re-zero process, but can also be adopted in a process of stopping the magnetic head 16 in a neighboring cylinder when a seek error occurs. The magnetic head 16 is relatively fast. Suitable for speed detection when moving.

  According to the present invention, the phase margin of the control system can be ensured. When performing only about 10 to 20% of the computation that requires the demodulation result of the current sample in the filter calculation during the period from the position demodulation to the output of the current value for the VCM, the computation delay And the effect of reducing the operational delay increases as the filter order n increases. On the other hand, when the prediction error is regarded as a kind of disturbance and a feedback path for feeding back the prediction error is provided, the gain can be increased without wasting the phase margin of the control system.

  Also, when calculating the speed of the magnetic head based on the phase difference of the EVEN layer in the position area, the correct speed can be determined quickly even at a relatively high speed, and the success rate of rezero processing etc. can be improved. Can do.

In addition, this invention also includes the invention attached to the following.
(Supplementary note 1) A signal reproduction method for reproducing servo information recorded on a recording medium using a head,
A matrix calculation step for performing current calculation regarding current and observer calculation regarding the position and velocity of the head on the recording medium from the reproduced servo information;
A filter calculation step for calculating a current value for driving the head based on a calculation result of the matrix calculation step;
The filter calculation step performs only an operation requiring the demodulation result of the current sample during the first period from the demodulation of the position of the head to the output of the current value, and the demodulation result of the past sample. The signal reproducing method is characterized in that an operation that requires is performed during a second period other than the first period.
(Supplementary note 2) The signal regeneration method according to supplementary note 1, wherein the filter calculation step performs a vector shift process that requires a past sample during the second period.
(Supplementary note 3) The signal regeneration according to Supplementary note 1 or Supplementary note 2, wherein in the matrix calculation step, the current calculation is performed during the second period, and the observer calculation is performed during the second period. Method.
(Supplementary note 4) The signal reproduction method according to any one of supplementary notes 1 to 3, wherein the second period is provided at a timing before and / or after the first period. .
(Supplementary Note 5) A signal reproduction method for reproducing servo information recorded on a recording medium using a head,
Subtracting the estimated position of the head from the actual position of the head on the recording medium obtained from the reproduced servo information to obtain a prediction error;
Feeding back the prediction error to a control system for calculating a current value for driving the head.
(Supplementary note 6) The method further includes the step of concentrating feedback of the prediction error on the first half of the current multirate output during multirate control in which the current value is output in a plurality of times. Signal reproduction method.
(Supplementary Note 7) A signal reproduction method for reproducing servo information recorded on a recording medium using a head,
A position demodulation step of demodulating phase information included in the servo information to demodulate the actual position of the head on the recording medium;
And a speed acquisition step for determining the speed of the head based on the phase difference of the demodulated phase information.
(Supplementary note 8) The signal reproduction method according to supplementary note 7, wherein in the velocity acquisition step, the velocity is obtained based on the phase difference only when the velocity of the head is equal to or greater than a certain value.
(Supplementary note 9) A signal reproduction method for reproducing servo information recorded on a recording medium using a head,
A position demodulation step of demodulating phase information included in the servo information to demodulate the actual position of the head on the recording medium;
A first speed acquisition step for determining the speed of the head based on the phase difference of the demodulated phase information;
A second speed acquisition step for determining the speed of the head based on the current and previous demodulated positions;
Adopting the speed obtained in the first speed acquisition step for a case where the speed is equal to or higher than a constant speed, and adopting the speed obtained in the second speed acquisition step for a case where the speed is smaller than the constant speed. A signal reproduction method characterized by the above.
(Supplementary Note 10) Reproducing means for reproducing servo information recorded on a recording medium using a head;
Matrix calculation means for performing current calculation regarding current and observer calculation regarding the position and velocity of the head on the recording medium from the reproduced servo information;
Filter calculation means for calculating a current value for driving the head based on the calculation result of the matrix calculation means;
The filter calculation means performs only a calculation requiring a demodulation result of the current sample during a first period from the demodulation of the position of the head to the output of the current value, and a demodulation result of the past sample. The storage device is characterized in that the calculation that requires is performed during a second period other than the first period.
(Additional remark 11) The said filter calculation means performs the vector shift process which requires a past sample during the said 2nd period, The memory | storage device of Additional remark 10 characterized by the above-mentioned.
(Supplementary note 12) The storage device according to Supplementary note 10 or Supplementary note 11, wherein the matrix calculation means performs the current calculation during the second period and performs the observer calculation during the second period. .
(Supplementary note 13) The storage device according to any one of Supplementary note 10 to Supplementary note 12, wherein the second period is provided at a timing before and / or after the first period.
(Supplementary Note 14) Reproducing means for reproducing servo information recorded on a recording medium using a head;
Means for subtracting the estimated position of the head from the actual position of the head on the recording medium obtained from the reproduced servo information to obtain a prediction error;
A storage device comprising feedback means for feeding back the prediction error to a control system for calculating a current value for driving the head.
(Additional remark 15) The control part which concentrates the feedback of the said prediction error on the first half part of the current multi-rate output at the time of the multi-rate control which outputs the current value divided into a plurality of times is provided, 14. The storage device according to 14.
(Supplementary Note 16) Reproducing means for reproducing servo information recorded on a recording medium using a head;
Position demodulating means for demodulating phase information included in the servo information in order to demodulate the actual position of the head on the recording medium;
A storage device comprising speed acquisition means for determining the speed of the head based on a phase difference of demodulated phase information.
(Additional remark 17) The said speed acquisition means calculates | requires speed based on the said phase difference, only when the speed of the said head is more than fixed, The memory | storage device of Additional remark 16 characterized by the above-mentioned.
(Supplementary Note 18) Reproducing means for reproducing servo information recorded on a recording medium using a head;
Position demodulating means for demodulating phase information included in the servo information in order to demodulate the actual position of the head on the recording medium;
First speed acquisition means for determining the speed of the head based on the phase difference of the demodulated phase information;
Second speed acquisition means for determining the speed of the head based on the current and last demodulated positions;
A speed obtained by the first speed acquisition means is employed when the speed is equal to or higher than a certain speed, and a speed obtained by the second speed acquisition means is employed when the speed is smaller than the constant speed. A storage device.
(Supplementary note 19) A head control method for reproducing servo information recorded on a recording medium,
Subtracting the estimated position of the head from the actual position of the head on the recording medium obtained from the reproduced servo information to obtain a prediction error;
A first feedback step of adding back the inputted predicted value and feeding back to a control system for calculating a current value for driving the head;
A second feedback step of feeding back to the control system only through a multiplier that multiplies the prediction error by a predetermined multiplication coefficient;
A control method comprising: an addition step of adding the first feedback step, the output value, and the output value of the second feedback step.
(Supplementary note 20) The supplementary note 19, further comprising the step of concentrating the feedback of the prediction error on the first half of the current multirate output during multirate control in which the current value is output in a plurality of times. Control method.
(Appendix 21) Means for subtracting the estimated position of the head from the actual position of the input head to obtain a prediction error;
First feedback means for adding the input predicted value and feeding back to a control system for calculating a current value for driving the head;
Second feedback means for feeding back to the control system only through a multiplier for multiplying the prediction error by a predetermined multiplication coefficient;
A control circuit comprising: an adding means for adding the first feedback means, the output value, and the output value of the second feedback means.
(Additional remark 22) The control part which concentrates the feedback of the said prediction error on the first half part of the current multi-rate output at the time of the multi-rate control which outputs the current value divided into a plurality of times, is further characterized by the above-mentioned 21. The control circuit according to 21.
(Supplementary Note 23) Reproducing means for reproducing servo information recorded on a recording medium using a head;
Means for subtracting the estimated position of the head from the actual position of the head on the recording medium obtained from the reproduced servo information to obtain a prediction error;
First feedback means for adding the input predicted value and feeding back to a control system for calculating a current value for driving the head;
Second feedback means for feeding back to the control system only through a multiplier for multiplying the prediction error by a predetermined multiplication coefficient;
A storage device comprising: the first feedback means, an output value, and an adding means for adding the output value of the second feedback means.
(Additional remark 24) The control means which concentrates the feedback of the prediction error on the first half part of the current multi-rate output at the time of the multi-rate control which outputs the current value divided into plural times, 24. The storage device according to 23.

  Although the present invention has been described with reference to the embodiments, it is needless to say that the present invention is not limited to the above-described embodiments, and various modifications and improvements can be made within the scope of the present invention.

It is a block diagram which shows 1st Example of the memory | storage device which becomes this invention. It is a figure explaining the servo information on a magnetic disc. It is a timing chart explaining a calculation delay. It is a flowchart explaining the principal part of the process which a reproduction | regeneration system performs. It is a figure explaining the shift process of the past sample. It is a flowchart explaining filter calculation. It is a functional block diagram explaining operation | movement of the principal part of 2nd Example. It is a figure which shows the principal part of the modification of 2nd Example. It is a timing chart explaining operation | movement of the modification of 2nd Example. It is a functional block diagram which shows the principal part of the modification of 2nd Example. It is a flowchart explaining a rezero process. It is a figure explaining an STW pattern. It is a figure which shows the recording pattern of the phase information for position demodulation. It is a figure explaining the state of zero phase difference in case the speed of a magnetic head is zero. It is a figure explaining the state of phase difference zero when the speed of a magnetic head is not zero. It is a flowchart explaining operation | movement of 3rd Example.

Explanation of symbols

1 MCU
2 DSP
3 User logic circuit 4 VCM control circuit 7 Servo demodulation circuit 11 VCM
15 Magnetic disk 16 Magnetic head

Claims (6)

A head control method for reproducing servo information recorded on a recording medium,
A calculation step of calculating an actual position of the head on the recording medium obtained from the reproduced servo information ;
Subtracting the estimated position of the head from the actual position of the head calculated in the calculating step to obtain a prediction error;
A first feedback step of feeding back the prediction error using a first feedback path to a control system that calculates the current value for driving the head by adding the prediction error ;
Using a second feedback path for feeding back the prediction error, a second feedback step of feeding back to the control system only through a multiplier that multiplies the prediction error by a predetermined multiplication coefficient;
In the control system, see contains an adding step for adding the output value of the output value and the second feedback step of the first feedback step,
A control method, wherein the gain is increased without wasting a phase margin of the control system by using the second feedback path different from the first feedback path . In the second feedback step , feedback of the prediction error using the second feedback path is concentrated on the first half of the current multirate output during multirate control in which the current value is output in a plurality of times . The control method according to claim 1, wherein: Means for subtracting the estimated position of the head from the actual position of the input head to obtain a prediction error;
A first feedback means for feeding back the prediction error using a first feedback path to a control system that calculates the current value for driving the head by adding the prediction error ;
Second feedback means for feeding back to the control system only through a multiplier that multiplies the prediction error by a predetermined multiplication coefficient using a second feedback path for feeding back the prediction error;
In the control system, comprising: an adding means for adding the output value of the first feedback means and the output value of the second feedback means ;
A control circuit characterized in that a gain is increased without wasting a phase margin of the control system by using the second feedback path different from the first feedback path .   The control unit according to claim 3, further comprising a control unit that concentrates feedback of the prediction error on a first half portion of a current multirate output during multirate control in which the current value is output in a plurality of times. Control circuit. Reproducing means for reproducing servo information recorded on the recording medium using a head;
Calculating means for calculating the actual position of the head on the recording medium from the reproduced servo information ;
Means for subtracting the estimated position of the head from the actual position of the head calculated by the calculating means to obtain a prediction error;
A first feedback means for feeding back the prediction error using a first feedback path to a control system that calculates the current value for driving the head by adding the prediction error ;
Second feedback means for feeding back to the control system only through a multiplier that multiplies the prediction error by a predetermined multiplication coefficient using a second feedback path for feeding back the prediction error;
In the control system, comprising: an adding means for adding the output value of the first feedback means and the output value of the second feedback means ;
A storage device , wherein the gain is increased without wasting a phase margin of the control system by using the second feedback path different from the first feedback path . The second feedback means concentrates the feedback of the prediction error using the second feedback path in the first half of the current multirate output during multirate control in which the current value is output in a plurality of times. 6. The storage device according to claim 5, further comprising control means.

Images