JPH0721708A - Method and device for controlling disk device - Google Patents

Abstract

PURPOSE:To calculate a required control current with simple calculation by using only rough positional information incorporated in servo information read out from an intermediate track placing between a present track and a target track and estimating a present speed of a head in a disk device. CONSTITUTION:A number of a desirable track is supplied to one of a subtracter 105, and a present track position by a sample is supplied to the other input of the subtracter 105. A positional error signal is generated by subtracting the present position by the sample from the desirable track position by the subtracter 105. The obtained error signal is supplied to a speed profile generator 110, and a speed command signal showing a desirable speed dealing with a positional error is supplied to the subtracter 115 by the generator 110. A desirable speed orbit is decided so that the head is moved rapidly in a place where the positional error is large, and the speed of the head is decreased as the positional error becomes small, and the speed of the head becomes zero at the time of arriving at the desirable track.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disc drive control method and control device, and more particularly to a digital control system used in a disc drive. More specifically, the present invention refers to a velocity estimator and a position estimator used during seek operations of a read / write head in a disk drive.

[0002]

2. Description of the Related Art Generally, a disk device has a head used for writing data in and reading data from a hard disk. In this case, the hard disk is configured to store data on a plurality of concentric tracks. Each of these tracks has a plurality of bits of consecutively arranged data. The data stored in the above hard disk includes servo data that is recorded in one track at a predetermined interval. Such servo data provides position data as used by a controller within the disk drive. The above servo data is called embedded servo data. This type of disk can store data magnetically, optically, or magneto-optically.

The head is connected to the actuator via an arm. This actuator is usually driven by a voice coil motor. A control current is supplied to the voice coil motor by the controller. This control current moves the arm so as to adjust the radial position of the head with respect to the disk. The head, arms, actuators and associated mechanical elements may conveniently be considered a set of controlled devices (Plant).

To increase the data storage capacity of a disk drive, it is also possible to use a large number of disks arranged vertically along a spindle. A number of heads corresponding to these disks are connected to the actuator via respective arms having a comb structure. In some control systems, only position information is recorded on a single disc called a "servo disc". If both the dedicated servo surface and the embedded servo data are used, the system of the servo part of the disk drive is "hybrid system (Hybrid system).
System) ". Servo disks are not provided in relatively modern systems. That is,
In such a system, the position information is obtained based on only a plurality of bits which form the embedded servo data arranged in the track of each disk.

During the seek operation, the controller changes the control current of the voice coil motor so as to move the head from the present track where the head is currently located toward the desired track (target track).
Further, the controller determines a value corresponding to the control current based on the position error signal indicative of the difference between the current track and the desired track.

[0006] More specifically, the position error signal presents an index to a velocity table that indicates the velocity trajectory, ie, the relationship between the desired velocity and the distance to travel. In this case, the control current is determined based on the difference between the desired speed of the head and the current speed. Furthermore, the control command, and thus the control current, includes a bias, that is, a term for compensating an external force.

In most modern disk drives, head speed is not used as the quantity to be measured. However, on the other hand, in this type of disk device, the position measurement value based on the servo data recorded in the track can be used as the amount to be measured while the head moves during the seek operation. Further, the current speed of the head can be estimated from the position measurement value obtained from the servo data.

In this case, since the servo data is recorded at a predetermined interval, the interval between continuously existing position measurement values is conveniently called a sampling interval (sampling time interval). There is. The controller operates in real time. That is, one control current value must be determined for each sampling interval.

In determining the control current, it is desirable that it takes only a short time so that it is supplied with a minimum delay. In other words, it is desirable to minimize possible errors in supplying the control current based on old measurements, ie measurements other than instantaneous measurements. Furthermore, it is desirable that the determination of the control current requires only simple calculations so that the processing power required of the controller is minimized. In other words, it is desirable to make the calculations so simple as to allow the use of relatively inexpensive controllers and to allow such controllers to control other tasks.

However, known (conventional) disk drives that use only embedded servo data and use the estimator approach need to use a relatively complex approach to position and velocity estimation. Therefore, the controller and the estimator take a relatively long time to determine the value of the control current at each sample (sample position). As a result, the controller and estimator described above are less likely to be utilized for other tasks.

Furthermore, known disk drives require complex techniques for compensating for errors in the measured precise position signal during head seek operations. One method of this kind is to write servo data by a special method at the time of manufacturing a disk. Another method is to use a software correction algorithm. Both approaches require complex processing and a relatively long time for this processing to complete.

The present invention has been made in view of the above problems, and provides a disk device capable of calculating a control current required for performing a seek operation of a head in the disk device by a simple calculation. It is a first object to provide a control method and a control device. further,
A second object of the present invention is to provide a control method and a control device for a disk device that can reduce the time required when the current speed of the head is estimated by the controller.

A third object of the invention is also to simplify the steps and means used by the controller to estimate the current speed of the head.

[0014]

In order to achieve the above object, in the control method and the control device for a disk device of the present invention, pre-written information including rough position information and precise position information is written. The seek operation of the movable head in a disk device of the type including a disk having a plurality of tracks having embedded servo information is controlled without using precise position information.

More specifically, according to the control method and the control apparatus of the present invention, a position error indicating the difference between the target track targeted by the head and the current track on which the head is located is generated. Furthermore, the speed command is presented as a function of this position error. Further, a speed error indicating a difference between the speed corresponding to the speed command and the value estimated with respect to the actual speed of the head is generated. further,
Without using the precise position information included in the servo data (servo information) read from the intermediate track located between the current track and the target track, the rough position information included in the servo information is used to The estimated speed is generated.

In this case, during the seek operation, the moving speed of the head is controlled by using the rough position information of the intermediate track located closer to the target track, so that the seek operation is mostly performed. The moving speed of the head can be set to a desired value without using precise position information in the servo data. Thus, in the present invention, when the head moves toward the target track, only the rough position information included in the servo information read from the intermediate track located between the current track and the target track is used. By estimating the current speed of the head in the disk device, the control current required to execute the seek operation of the head can be calculated by a simple calculation, so that the head can be moved quickly toward the target track in a short time. It becomes possible to move.

[0017]

The invention will be readily understood by the following description of specific embodiments illustrated by the accompanying drawings (FIGS. 1 to 19). Embodiments of the present invention and related matters will be described in detail below with reference to the accompanying drawings. In this case, the same constituent elements existing in different drawings are denoted by the same reference numerals.

FIG. 1 is an enlarged view showing a part of a disk used in the present invention. Here, a portion of a disk provided in a disk device to which the present invention is easily applied is illustrated. This disc is composed of a plurality of tracks concentrically arranged on it. Although common in the art, the term cylinder is sometimes used instead of truck. This cylinder is composed of a plurality of tracks that are arranged at the same position on a plurality of disks that are stacked on the spindle in the disk device.

In a commercially available disk device, each track (for example, k to k + 6) is composed of a plurality of bits arranged continuously. Each of these multiple bits is recorded in a physical portion that exists in a discrete state on the disc. For example, in a magnetic disk device, the width of each track is about 0.00037 inches (about 0.0094 inches).
mm), and the recording density is about 67,000 bits / inch (2,638 bits / mm). As a result, the area per bit is approximately 370 micro inches (9.40 μm
) × 15 micro inches (0.381 μm).

Here, the data is recorded on the track in units of segments. For example, 54 segments are included for one track. Each segment begins with a portion of a servo frame, which is followed by a portion of data. The former part of the servo frame is recorded when the disc is manufactured, while the latter part of the data contains useful data, eg user data. In an embodiment of the invention, the disc typically rotates at a speed of 4,500 rpm. As a result,
One segment has a time length of about 250 microseconds (μsec) and a servo frame has a time length of about 35 microseconds. The bits included in the servo frame are called servo bits and form servo data.

The servo sampling interval is such that it starts with the reading of a servo frame and ends just before the reading of the next servo frame. FIG. 2 is a format diagram showing data items included in the servo frame in the disk of FIG. The first approximately one-third of a servo frame contains the bits used to capture automatic gain control (usually abbreviated as AGC, as also shown in FIG. 2) information. The set of bits next to the AGC information acquisition bit represents a servo mark. This servo mark is used as a synchronization reference for decoding the servo information corresponding to the servo data that arrives immediately after, and also as a timing reference for speed control for the servo frame, that is, disk rotation control. It is what is done. Further, the subsequent set of bits represents the servo frame number recorded by the Gray code. These bit sets are mainly used for the diagnostic operation of the disk device and are not related to the present invention. After that, there is a bit idle period generally called a space (usually abbreviated as SP as shown in FIG. 2). further,
The continuous set of bits present after this bit pause period represents servo burst data.

[0022] The servo burst data is typically, N is recorded as quadrature pattern ', N', the lower portion of the bar Q Oyo <br/> beauty Q (N 'and Q in FIG. 2 As shown, it should be added to the upper part of N'and Q originally, but it is difficult to add a bar to the upper part of characters such as N'and Q in the current electronic filing system, so that the present specification For the sake of convenience, it includes four values such as those shown in N'and Q below the bar). These four values together provide precise position information. The servo burst data described above allows for the determination of the actual position of the head located between the ideal track centerlines. The head position is determined by the resolution defined by an analog / digital converter (hereinafter abbreviated as A / D converter) provided in the controller of the disk device. This A / D converter is for digitally displaying the head position.

The servo burst data described above typically consists of 8 bits and presents a resolution of ± 1/2 track. The present invention does not use servo burst data during most of the seek operation. Instead, the servo burst data is used only in the final period of the seek operation and the track following operation. That is, servo burst data is used when the head is located in the immediate vicinity of the desired track. As described above, according to the present invention, good seek performance can be obtained without using servo burst data indicating precise position information. Therefore, a special method conventionally used as a matter of course is used. Eliminates the need to use

Other techniques for providing precise position information are well known to those skilled in the art. Therefore, the term "fine position information" as used herein encompasses all the other approaches mentioned above. The last part of the servo frame is a set of bits indicating the track address or cylinder address, and is therefore also called coarse position information. In this case, the track address is
It is represented by the Gray code and has a sufficient length. In other words, it contains bits of sufficient length to identify each track on the disc as unique.

Hereinafter, when a gray code is referred to in this specification, this gray code is intended to mean a track address. By reading the bits of the track address during the seek operation, the actual track on which the head is located can be determined with a resolution of ± 1/2 track. Figure 3
FIG. 3 is a diagram showing a speed control loop used in an apparatus for controlling a disk device of the present invention.

As shown in FIG. 3, the part of the speed control loop includes the subtracters 105 and 115 and the speed profile generator 1.
10, a gain circuit 120, adders 125 and 145,
Digital / analog converter (indicated as DAC in FIG. 3: hereinafter referred to as D / A converter in the text) 13
0 and the transconductance amplifier 132 (G in FIG.
And a set of devices to be controlled (sometimes referred to as a set of devices to be controlled) 135. further,
The part of the speed control loop of FIG.
0, the precision position demodulation device 150, the determination circuit 155,
Speed / position estimator 160 and three switches 165, 1
70 and 175 and a current sensor 180.

In FIG. 3, a desired track identifier, also called the target track, eg, track number, is provided to one input of subtractor 105. The current track position by sample is provided to the other input of subtractor 105. The subtractor 105 operates to subtract the current track position by sample from the desired track position. The position error signal e is generated by the subtraction operation of the subtractor 105.

The position error signal e thus obtained
Are supplied to the velocity profile generator 110. The velocity profile generator 110 operates to supply to the subtractor 115 a velocity command signal indicating a desired velocity of the head corresponding to the above position error. In one embodiment, such position error serves as an indicator for a look-up table that indicates a desired velocity trajectory for the head of the disk drive to follow during a seek operation. The desired velocity trajectory is set so that the head moves as fast as possible if the position error is large. On the other hand, such a desirable velocity trajectory is determined so that the velocity of the head decreases as the position error decreases, and the velocity of the head becomes zero (0) when the desired track is reached. Alternatively,
The velocity trajectory is defined such that the velocity of the head is zero (0) just before the desired track to avoid head overshoot.

In another embodiment, velocity profile generator 110 performs the calculations to determine the velocity command. In yet another embodiment, the velocity profile generator 110 uses a look-up table or a velocity command calculation process to generate the velocity command signal. In addition, the current speed estimate is also provided to the subtractor 115. The subtractor 115 operates to subtract the estimated current speed from the desired speed. The subtractor 115 described above operates to supply the speed error signal obtained as a result of such subtraction operation to the gain circuit 120.

Further, the gain circuit 120 operates so as to adjust the bandwidth of the entire speed loop (speed control loop) by adjusting the amplitude of the speed error signal.
The gain circuit 120 supplies the speed error signal thus adjusted to the adder 125. Signal EXT represents the result of adding the bias force compensation term to the open loop feedforward compensation term. This signal EXT is also supplied to the adder 125. This adder 125
It operates so as to add the adjusted speed error signal to the signal of the compensation term. A control current signal U (z) is generated by the adding operation of the adder 125. This control current signal U
(Z) presents different values for each sampling interval.

Further, the control current signal U (z) is D / A
It is supplied to the converter 130. This D / A converter 130
After converting the control current U (z) into an analog signal,
The transconductance amplifier GM has a transconductance value GM (in FIG. 3, the block of the transconductance amplifier 132 is indicated as GM). This transconductance amplifier 132 converts the above control current signal into a control current and completes a set of devices to be controlled 1.
35. In this set of devices 135, the actuator assembly responds to the control current by moving the head in its radial direction relative to the disk. In this case, the actuator assembly has a voice coil motor and a rotary actuator with a magnetic head. This magnetic head is a suspension (Suspension) at the end of the actuator arm.
Fixed on top.

When the head moves, it is located below the head and reads such servo data as it rotates the servo data previously written to the disk. Based on the data read by the head, the device set 135 to be controlled generates a physical position signal including a servo mark, a servo burst, and a gray code track address represented by the servo data. This physical position signal is generally supplied to the position demodulation device 140 and the fine position demodulation device 150.

The position demodulator 140 generally operates to convert a physical position signal into a digital format signal. The conversion operation of the position demodulator 140 is performed by using the servo mark included in the servo data read by the head as a clock for the conversion operation. Further, the above-mentioned general position demodulation device 140
Operates to extract the track number from the digitized physical position signal. The track number extracted in this manner is added to the adder 14 as a rectangular position signal.
5 is supplied.

The fine position demodulator 150 operates to extract precise position information based on servo burst data obtained from a physical position signal. Further, the precision position demodulation device 150 supplies a precision position signal corresponding to precise position information to the switch 165. The switch 165 supplies its output to the adder 145. The decision circuit 155 operates to determine rough position information, that is, whether the quadrature position signal is used alone, or whether the rough position information is used in combination with the fine position information. In the embodiment of the present invention, whether to combine rough position information and precise position information is
Depends on the distance from the target track. Precise position information is used only if the speed of the head is below a predetermined threshold and it is guaranteed that the accurate position information is valid. More specifically, when the distance of the head from the target track is smaller than 20 tracks, the determination circuit 155 controls the switch 165 so that precise position information is supplied to the adder 145. Alternatively, in another alternative, the criteria for using the fine position information can be a function of the current velocity.

The adder 145 operates so as to generate a sampled position signal X1 (z) by combining the rough position information with the precise position information selectively supplied via a switch or the like. . Such a position signal X1 (z) is also called a composite position signal, that is, a composite position signal for each sampling interval.

The control current signal U (z) is supplied to the switch 175.
It is also supplied to one terminal of. This switch 175
During the deceleration phase of the seek operation, the speed / position estimator 16
0 to supply the control current signal U (z). The composite position signal X1 (z) is also provided to the velocity / position estimator 160. This velocity / position estimator 160
Is a signal X1ES showing the estimated value of the current position of the head.
T (z) and a signal X2EST indicating the estimated speed value
Operates to generate (z). The former position estimate is called the expected position in the future, and the latter speed estimate is called the current speed estimate. These two types of estimated values represent the values at the time of each sampling, that is, the values after the read operation of each physical position. The above velocity / position estimator 1
60 is illustrated in greater detail in FIGS. 6 and 7 discussed below.

Furthermore, the decision circuit 155 switches the switch 170 so as to supply the signal X1EST (z) of the expected position or the composite position signal X1 (z) as the current sample position to the subtractor 105. Control. In general,
The composite position signal is selected as the current sample position signal. However, when the servo frame is erroneously read because the timing mark in the servo data cannot be found or the current position is considered to be incorrect, the determination circuit 155 determines that the expected position is Selected as the current sample position. For example,
If the current track at the start of the seek operation is “track 200” and the target track is “track 230”, the current position of “track 100” during the seek operation is considered incorrect.

Velocity / position estimator 160 includes a first order digital filter as will be described. This first-order digital filter becomes unusable when the head approaches the target track in response to the control signal from the determination circuit 155. When a first-order digital filter is used under such conditions, the above-mentioned disabled state occurs at at least one sampling after the fine position information is incorporated into the rough position information. As a result,
A possible transient response, such as that output from the velocity / position estimator 160 due to a spike in the control current signal U (z), can be prevented. Another preferred alternative is to increase the bandwidth of the first-order digital filter by multiplying it by a factor of about 5, after the precise position information has been incorporated into the coarse position information. For example, an attempt is made to widen the bandwidth of the digital filter by changing the gain constant of the primary digital filter in response to the control signal from the decision circuit 155.

By using this digital filter, the hard disk drive using the control method and control device of the present invention continuously changes during the entire seek operation even when only rough track information is used. It is possible to achieve almost the same seek performance as in the case of using information, that is, a combination of precise position information and rough position information.

Velocity profile generator 110, gain circuit 1
20, decision circuit 155 and velocity / position estimator 160
Can be implemented by software routines executed by an embedded controller such as, for example, an Intel 196KC integrated circuit chip. Alternatively, the velocity profile generator 110 described above,
The gain circuit 120 and the like, like an integrated circuit chip of 320C25 manufactured by TI Co., Ltd., is a software program executed by a control chip for digital signal processing, which is directed to perform a head positioning and rotation control function in a disk device. It can be realized by a routine.

During the acceleration phase of the seek operation, the control current signal U (z) should be as large as possible.
That is, it is desirable that the D / A converter 130 be set to the maximum value that can be displayed with the number of usable bits. However, due to temperature and power fluctuations,
The actual current supplied to the coil of the voice coil motor at the maximum level in the D / A converter 130 cannot be reliably predicted only from the control current signal U (z).

In order to eliminate such an inconvenience, during the acceleration stage of the seek operation, the estimated value of the speed (signal X2EST (z)) estimated by the speed / position estimator 160 from the start of the seek is shown. Value) reaches within 75% of the required speed as output from the speed profile generator 110 until the decision circuit 155.
Preferably controls switch 175 in the following manner.

That is, a signal presented at one terminal 175B of this switch 175 and proportional to the actual coil current in the voice coil motor (usually abbreviated as VCM) is presented at the other terminal 175A. The switch 175 is controlled so as to be supplied instead of the calculated control current signal U (z).

The current sensor 180 senses the actual coil current in the voice coil motor, and outputs a signal proportional to the sensed coil current to the terminal 1 of the switch 175.
It operates to supply to 75B. Current sensor 1 above
80 is, for example, a current sensing resistor, an amplifier and a D / A.
It is composed of a converter. 4 and 5 are views for explaining the seek operation according to the present invention, and FIG. 4 is a graph showing the relationship between the control current of the voice coil motor generated in the control device of the present invention and the sampling time interval. 5 is a graph showing the relationship between the head speed and the sampling time interval in the control device of the present invention.

Here, the case where the head is considerably separated from the target track will be described. Further, in FIG. 4, the control current signal U (n) (n is an arbitrary natural number) corresponding to the control current of the voice coil motor is displayed on the vertical axis of the graph, and in FIG. 5, the speed signal corresponding to the speed of the head is displayed. X2 (n) is displayed on the vertical axis of the graph. Note that, hereinafter, the control current signal U (n) of the voice coil motor is defined as the controller output, and the speed signal X2 (n) of the head is defined as the speed of the head.

More specifically, at each sampling time t, (N-1) T, (N-1) T,
At NT and (N + 1) T, where T represents one sampling interval, i.e. sampling period, physical position samples are presented as a result of the servo frames being read. The head speed is not physically sensed as shown in FIG. For example, at the time of sampling NT, the head speed is X2 (N).
Is. Further, as is apparent from FIG. 4, at the sampling time NT, the controller output is U (N-1).
Is. Sample-correction delay Δ (Δ = (1-
After m) · T) has elapsed, the inclination of the speed of the head changes. Here, the constant m will be described. During the remaining period mT of the sampling interval, that is, the sampling period T minus the sample-correction delay Δ, the controller output remains U (n), and
The head speed slope does not change. At the next sampling time (N + 1) T, the next servo frame is read,
The controller output adjustment process is repeated.

At each sampling (N-1) T, N
"N" used for T and (N + 1) and the like may be represented by a lower case "n" in the description of formulas and the like below for convenience. Then, the velocity / position estimator 1 of FIG.
60 will be described in detail. Here we use the z-domain notation, where z ^-1 represents the delay of one sampling interval.

In this case, U (z) is defined as the controller output, similar to U (n) described above. This U (z) is a value that corresponds to the current supplied to the voice coil of the actuator and also corresponds to the acceleration of the voice coil motor represented by track / sample ² .
It is expressed in units of the least significant bits (usually abbreviated as LSB) in the output of the D / A converter 130.

X1 (z) is defined as a sample position and is expressed in track units. X1EST (z) is
It is defined as an estimate of the current position of the head based on previous information and is expressed in track units. X2EST
(Z) is defined as an estimate of the current velocity of the head based on the sample position and previous information and is expressed in track / sample units.

Thus, U (z), X1 (z), X1
When EST (z) and X2EST (z) are defined, X1EST (z) and X2EST (z) are respectively represented by formula (1) and formula (2) as shown in the following [Equation 1]. .

[0051]

[Equation 1]

In this case, by expressing the head speed in the unit of track / sample and the controller output in the unit of track / sample ² , the above equation (1) is obtained.
And the sampling period is excluded from equation (2). Derivation of the above equations (1) and (2) will be described below in connection with the display of these equations by the difference equation. It will be readily appreciated that, as mentioned above, the signal U (z) represents the coil current in the voice coil motor, rather than the required control current signal.

The constants C0, C1, D1, G1, K1 and K2 provide sufficient performance for the velocity / position estimator and are compatible with the dynamics of the set of devices being controlled (actuator assembly, etc.). Is a constant that is selected as follows. Further, the constants C0 and C1 are filter constants having a relationship of C1 = 1−C0. These constants C0, C
1 is when only rough position information of quadrature is used,
It is selected to ensure a bandwidth of about 300 Hz. On the other hand, if only precise position information is used, it is about 1
It is selected to ensure a bandwidth extended to higher frequencies, such as ~ 2 kHz. In one embodiment of the invention, the constant C0 is about 0.35.

Furthermore, the constant D1 is adjusted as a function of the control system sample rate, the velocity / position estimator bandwidth, and other parameters. In one embodiment of the invention, the constant D1 is about 2.5. Further, the constant G1 is a feedback correction constant. This constant G1 is set to 0..0 while considering the simulation result and the experimental result so that the estimated value of the velocity obtained based on the predicted position derived from the composite position information can be corrected. It is set to a value between 5 and about 0.7.

Further, the constants K1 and K2 are selected so as to allocate the controller output in a predetermined distribution with respect to the sampling period. In this case, various variables are set as follows. DAC is the gain of the D / A converter 130 (in FIG. 3, the block of the D / A converter 130 is shown as DAC) and is represented by the voltage value for the least significant bit (LSB). .

GM is the transconductance gain of the transconductance amplifier 132 for driving the voice coil (in FIG. 3, the block of the transconductance amplifier 132 is indicated as GM), and ampere / volt (A /
V). KT is the torque constant of the voice coil motor and is expressed in Newton meters / ampere (Nm / A).

J is the inertia of the head arm assembly and is represented by Kg · m ² . -TPI is a value of tracks per inch of the disc. KP is a constant that is calibrated at power on,
It also shows the acceleration constant of the set of devices to be controlled. Specifically, it is expressed as in the following [Equation 2].

[0058]

[Equation 2]

Where T is the sampling period (described above), Δ is the sample-correction delay,
m is defined as m = (T−Δ) / T, and is a constant having a value of about ½ in one embodiment of the present invention.

Further, K1 = 0.5.KP.m and K2 = 0.5.KP. (1-m) are set. The nominal value and the worst value of the acceleration constant KP are calculated from the design values of the disk device. This disk drive is calibrated at power-on due to manufacturing dimensional tolerances and / or temperature variations. More specifically,
This calibration is performed by adaptive control during the operation of the disk device and as part of the factory calibration process during manufacturing. Further, the above calibration involves accelerating the actuator using a constant current for a defined period of time, and measuring the distance traveled by the head in response to this acceleration. Including. Further, the acceleration constant is calculated based on the measured distance. In one embodiment of the invention, the constant KP is about 0.015 tracks /
Sample ² / LSB.

By selecting the above constants correctly,
Even under adverse conditions, such as miscalibration of the acceleration constant KP of the set of controlled devices, or miscalibration of the bias force acting on the head arm assembly,
Estimated value of current speed of head (estimated speed) X2E
ST (z) is a value that accurately estimates the actual speed of the head. In other words, the control system of the present invention is robust. G1 · (X1 (z) −X1EST (z)) which is the feedback correction term in the present invention
Plays an important role in the robustness of the velocity / position estimator. This is because the above feedback correction term prevents an error from occurring due to a change in actuator characteristics that is different from the values used for the purpose of setting various constants.

Further, the estimated value of the present position of the head (estimated position) X1EST (z) is a value that accurately predicts the true position of the head at the next sampling under all important dynamic conditions. become. Some of the important kinetic conditions mentioned above include acceleration periods, areas of constant velocity, deceleration periods, and periods in which the value of deceleration changes. More specifically, the head positioning system using the control method and the control device of the present invention can not only withstand the step function of the controller output but also recover the original state. This type of recovery operation can typically be performed in a period within two sampling cycles.

FIG. 6 is a block diagram illustrating one embodiment of a velocity / position estimator that may be used in the velocity control loop of FIG. As can be seen by considering FIG. 6, the velocity / position estimator 160 of FIG. 6 is a direct implementation of equations (1) and (2) above. Further, in FIG. 6, velocity / position estimator 160 includes input terminals 205 and 210, delay circuits 215, 225 and 270 including delay elements, subtractors 220 and 235, and adders 245, 255 and 28.
0 and the multipliers 230, 240, 250 and 275,
The switch 260 and the output terminals 265 and 285 are provided.

The composite position signal X1 (z) is input to the input terminal 2
05, and further is supplied from this input terminal 205 to the delay circuit 215. This position signal X1 (z) is
At the same time, it is supplied to the subtractor 220 and the subtractor 235. The delay circuit 215 outputs the delayed position signal z ⁻¹ · X1.
It operates to supply (z) or X1 (z-1), in other words the position signal X1 (z) obtained from the previous sampling interval, to the adder 280 and the subtractor 220.

The control current signal U (z), that is, the control command is supplied to the input terminal 210 and further supplied from this input terminal 210 to the delay circuit 225. The delay circuit 225 has a stored value indicating a control command in the sampling interval two before. Further, the delay circuit 225 multiplies the previous control command value z ⁻¹ · U (z) by the constant K1 and also adds the constant K2 to the control command value z ⁻² · U (z) at the sampling interval two before. Multiply,
Further, it operates so as to add the values thus multiplied. Further, the delay circuit 225 uses the result (addition result) obtained as the command signal belonging to the past in the multiplier 23.
0 and adder 280.

The velocity signal z ⁻¹ · X2EST (z) estimated from the previous sampling interval is supplied from the delay circuit 270 to the adder 280. The adder 280 operates to add a plurality of types of signals supplied to the adder itself. These plural kinds of signals are the delayed composite position signal z ⁻¹ · X1 (z), the velocity signal z ⁻¹ · X2EST (z) estimated from the previous sampling interval, and the command belonging to the past. Signal K1 ・ z ^-1・ U (z) + K
Includes 2 · z ⁻² · U (z). Furthermore, the adder 28 described above
0 operates to provide the expected position signal to output terminal 285 and subtractor 235.

As can be seen from the above description, the expected value of the position signal X1EST (z) depends on the measured value up to and including the previous sample. Franklin, Powel
l) and Workman methods {Digital Control of Dyn
amic Systems), 2nd Edition, Addison Wesley (Ad
dison Wesley), published 1990, see §6.3}, the signal X1EST (z) is the value expected by the estimator.

More specifically, the nth sampling interval (strictly speaking, it should be expressed as “Nth sampling interval” selected from arbitrary n, but here, for simplification of explanation, , "Nth sampling interval").
If (z) is supplied to a controlled set of devices, ie the set of controlled devices, the next sampling interval before the next physical position sample is obtained from this controlled set of devices. Position signal X1 predicted by estimation for
EST (z) can be acquired quickly. In this way, the period of time until the expected position signal X1EST (z) is acquired is called the period of the provisional calculation. During the part of this provisional calculation, the command signals D1 * (K1 · z ⁻¹ ·) that are weighted and belong to the past
U (z) + K2 · z ⁻² · U (z) or the weighted and delayed estimated speed signal C1 · z ⁻¹ · X
2EST (z) is also obtained.

As a result, during the period other than the temporary calculation period, that is, during the in-line portion within the sampling interval, only two multiplication functions and two addition functions need to be performed. Due to this fact, a short sample-correction delay Δ can easily be realized. Although not discussed here, in one embodiment of the present invention, the sample-correction delay Δ is set to a value of about half the sampling interval, due to the presence of other necessary calculations.

The subtractor 235 outputs the composite position signal X1.
The expected position signal X1EST (z) is subtracted from (z), and the result obtained as the position difference data by this subtraction operation is supplied to the multiplier 240. Further, the multiplier 240 operates to multiply the position difference data by a constant G1 and supply the result obtained as a weighted position difference signal to the adder 245 by this multiplication operation. Here, the weighted position difference signal is also called a feedback correction signal.

The multiplier 230 operates so as to multiply the command signal belonging to the past by a constant D1 and to supply the result obtained as the command signal belonging to the past weighted by this multiplication operation to the subtractor 220. This subtractor 220
Operates to subtract the composite position signal obtained from the previous sampling interval from the composite position signal for the current sampling interval and add the result of the subtraction to the weighted and past command signals. By the series of operations of the subtractor 220, a set of signals to be supplied to the adder 245 is generated.

The adder 245 adds the above-mentioned group of signals and the weighted position difference signal, and supplies the result obtained as an intermediate signal by this addition operation to the input terminal 260A of the switch and the multiplier 250. To work. Further, the multiplier 250 operates to multiply the intermediate signal by the constant C0. The weighted intermediate signal as the result of this is the adder 2
55.

The switch 260 is the decision circuit 155 of FIG.
It is controlled by a control signal from. If the precise position information is incorporated into the coarse position information and no filtering is required, the weighted intermediate signal is the estimated speed signal X2EST (z), which is the input terminal 260A of the switch. From the output terminal 265.

The velocity signal z ⁻¹ · X2EST (z) estimated from the previous sampling interval is supplied from the delay circuit 270 to the multiplier 275. The multiplier 275 multiplies the speed signal z ⁻¹ · X2EST (z) described above by a constant C1, and further supplies the result obtained as a weighted and delayed estimated speed signal to the adder 255.

Delay circuit 270 and multipliers 275 and 25
0 includes a first order digital filter. Adder 255
Is the weighted and delayed estimated speed signal,
The weighted intermediate signal is added, and the addition result is supplied to the input terminal 260B of the switch.
For example, in the majority of seeks, if only coarse position information is used, output filtering is required. Further, in this case, the signal applied to the input terminal 260B of the switch is equal to the estimated speed signal X2EST.
(Z) is supplied to the input terminal 265 of the switch. Alternatively, as another example, the filtering process can be performed even when precise position information is used. In this case, a filter with a higher frequency range is used than if only coarse position information is used.

As can be seen from the above description, the value of the estimated velocity signal X2EST (z) depends on the measured value up to and including the previous sample. Using the Franklin, Powell and Workman method described above, the estimated velocity signal X2E
ST (z) is the current signal of the estimator. The velocity / position estimator of the present invention has a relatively simple structure. The reason for this is that each of the position and velocity estimates is relevant at a particular sampling time, and during the period after the sample-correction delay, i.e. when the command signal is applied, the above position and velocity estimates are There is no attempt to identify However, as mentioned above, the sample-correction delay involves the selection of the weighting constant.

By choosing the constant D1 correctly, the velocity / position estimator of the present invention does not need to estimate the value after this sample-correction delay, even if there is a large sample-correction delay. It is possible to present accurate results. FIG. 7 is a block diagram showing a modification of the speed / position estimator of FIG. Here, the velocity / position estimator 160B of FIG. 7 will be described in comparison with the velocity / position estimator 160 of FIG.

By substituting the equation (1) into the equation (2), the following equation (3) is calculated.

[0079]

[Equation 3]

The equation (3) thus obtained is simplified as the following equation (4) by collecting the terms for each constant.

[0081]

[Equation 4]

This equation (4) is calculated between consecutive position samples.
The sum of the differences between the
It represents the sum total. Furthermore, it is added in this way
Is a first-order low-pass filter (low-pass filter) with coefficient E.
Filter processing is performed. Sample-correction
The delay is factored into the constants K1 and K2. This
For K1z corresponding to command signals belonging to the past^-1
・ U (z) + K2 ・ z ^-2・ U (z) term is one sun
Similar to coil current for pulling period and X1
(Z) -z^-1-The term X1 (z) is a digital differentiator
Will be equivalent to. Complementary feedback in equation (2)
Positive term G1 · (X1 (z) -X1EST (z))
Is expanded in equation (3), and constants A and B in equation (4)
And E.

The above equations (1) and (4) are generally executed by the block diagram shown in FIG. The example of the velocity / position estimator of FIG. 7 has an advantage that the total sum of the calculation processing of the processor is smaller than that of the velocity / position estimator of FIG. 6 described above. Because, the speed of /
This is because the position estimator requires only a single multiplication operation, a single addition operation, and a single subtraction operation in the inline portion within one sampling interval.

In FIG. 7, velocity / position estimator 160B
Are input terminals 405 and 410 and delay circuits 415 and 43.
5 and 450, the subtractor 420, and the adders 430 and 4
60, multipliers 425, 440 and 450, and output terminals 445 and 465. This speed / position estimator 1
The components of 60B generally perform similar operations as for velocity / position estimator 160 of FIG. 6 above. Therefore, here, in order to simplify the explanation, the speed /
The description of the operation of each component of the position estimator 160B will be omitted.

During the portion of the in-line portion within the time critical sampling interval, subtractor 420, to produce the estimated velocity signal X2EST (z),
Only the multiplier 425 and the adder 430 need be operated. Other components in the velocity / position estimator 160B,
That is, the delay circuits 415, 435 and 450, the multipliers 440 and 450, and the adder 460 can be operated during the part of the provisional calculation within the sampling interval that is not time-critical.

In order to facilitate the description of the software execution example of the velocity / position estimator 160B, the above equations (1) and (2) are converted into the following equations (5) and (6), respectively. As shown in, the difference equation is rewritten.

[0087]

[Equation 5]

[0088]

[Equation 6]

The process of deriving these equations (5) and (6) will be described with reference to FIGS.
It will be described in detail with reference to FIG. A complete closed loop control system, including an estimator and controller, is designed with good and stable margins. That is, the closed loop control system described above is designed such that variations in the control current of the voice coil motor are suppressed when following a velocity profile of a type having a constant deceleration. As a result, the incremental change in acceleration during one sampling interval becomes small. For most of the seek, the acceleration does not change rapidly within one sampling interval. Therefore, a good estimate of the incremental change in position during each sampling interval is made by (average velocity) .multidot. (Sampling period). Based on this quote
A new position is calculated by the following equation (7).

[0090]

[Equation 7]

In the preferred embodiment, the transient response of the estimator, such as that which results from this change in acceleration, is such that it recovers within two sampling periods even when the acceleration changes abruptly. Has become. in this case,
Since the speed is expressed in units / sample, the sampling period T need not be included in the equation. Further, the equation (7) can be rewritten as the following equation (8).

[0092]

[Equation 8]

0.5 · (X2 (n + 1) -X2 (n))
The term represents the incremental change in velocity in a section as caused by acceleration during each sampling cycle. Further, X2 (n + 1) -X2 (n) is expressed by the following equation (9).

[0094]

[Equation 9]

By combining the equations (8) and (9), the following equation (10) is calculated.

[0096]

[Equation 10]

In the above equation (5), X2EST
Let (z) be X2 (z), and K1 = 0.5 · KP ·
When m and K2 = 0.5 · KP · (1-m), the equation (10) becomes equal to the equation (5). 8,
FIGS. 9, 10 and 11 are flowcharts No. 1, No. 2, No. 3, and No. 4 of the flowchart for explaining the operation of the velocity / position estimator shown in FIG. From these flow charts, the above equation (in particular,
A software routine for performing the difference equation) is illustrated.

8 to 10, steps 300 to
Step 365 includes the inline portion of the velocity / position estimator routine. That is, step 300 above
-Step 365 shows the calculation procedure of the estimated speed (signal) X2EST (n) based on Formula (6). On the other hand, step 370 of FIG. 10 and FIG.
~ Step 399 includes part of the provisional calculation of the velocity / position estimator routine.

First, in step 300, an A / D conversion operation is started by the A / D converter in the control system. Next, in step 305, a test is performed to determine if the precise location information should be used. If the control system does not meet the criteria for using precise position information, only coarse position information is used as the composite position. On the other hand, if it is determined by the test in step 305 described above that the precise position information should be used, in step 310, the precise position information is read and combined with the rough position information, A true composite position is generated.

In addition, in step 315, the composite locations are tested for validity, ie, the validity of the location sample. This check is performed, for example, by using hardware bits. If the composite position is not valid, then in step 320, the composite position is replaced by the current position estimate obtained during the tentative calculation within the previous (previous) sampling interval.

Here, the above equation (6) is expressed by the following [Equation 1]
It can be rewritten as in [1].

[0102]

[Equation 11]

Further, in step 325 of FIG. 9,
It will be readily appreciated that the value of a above is calculated by subtracting the sample of the previous position from the sample of the current position. Further, in step 330, it is easily recognized that the value of b is calculated by subtracting the estimated value of the current position obtained in the provisional calculation part within the previous sampling interval from the current position sample. Will be done. Furthermore, it will be readily appreciated that in step 335, the value of c above is calculated by a multiply operation (b * G1 = c).

Further, in step 340, in the portion of the provisional calculation (interruption) within the immediately preceding sampling interval,
D1 * (K1 · U (n-1) + K2 · U (n-2)) =
It will also be readily appreciated that the value of d above is calculated by calculating d in advance. Further, it will be readily appreciated that in step 345 the value of e above is calculated by adding the values of a, c and d. Furthermore, it will be readily appreciated that the value of f above is calculated by performing the multiplication C0 * e in step 350. further,
In step 355 of FIG. 10, the speed X2EST estimated in the provisional calculation part in the immediately preceding sampling interval is calculated.
By multiplying (n-1) by the constant C1, the above g
It will also be readily appreciated that the value of is calculated.
Furthermore, it is clear that in step 360 the current speed estimate X2EST (n) (f + g = X2EST (n)) is obtained.

Further, in step 365, the estimated value X2EST (n) of the current speed is passed to the control law routine, and the control law is executed for this estimated value X2EST (n). The resulting controller output U (n) is passed to the D / A converter as a correction value. The above control law includes the functions of the velocity profile generator 110, the subtractor 115, the gain circuit 120 and the adder 125 of FIG.

At step 370, the portion of the interim calculation within the sampling interval is started by obtaining the controller output U (n). Further, step 37
5, the term h (K1 · U) of the command signal belonging to the past
(N) + K2 · U (n-1) = h) is calculated. Further, in step 380 of FIG. 11, the estimated position value X1EST (n + 1) for the next sampling interval is calculated.
Is obtained. Further, step 385 and step 3
90 is multiplication D1 * h and C1 * X2ES, respectively
Perform T (n) and store these multiplication results. Further, in step 395, the values of X1EST (n + 1) and U (n) above are shifted in memory and the variables in this memory are updated to reflect the new sampling interval. Finally, the provisional calculation routine is
The process ends at step 399.

As in the above case, in order to facilitate the explanation of the software execution example of the velocity / position estimator 160B of FIG. 7, the above equations (1) and (4) are changed to
As shown in the following equations (11) and (12),
Let's rewrite it as a difference equation.

[0108]

[Equation 12]

In this case, for convenience of explanation, the above equation (6) is used.
The expression (11) having the same form as the above will be described repeatedly. FIGS. 12, 13 and 14 show the first, second, and third flowcharts for explaining the operation of the velocity / position estimator of FIG. 7, respectively. From these flow charts, the above equation (in particular,
A software routine for performing the difference equation) is illustrated.

Steps 500-565 of FIGS. 12 and 13 include the inline portion of the routine of velocity / position estimator 160B (FIG. 7). That is, the above steps 500 to 565 show the calculation procedure of the estimated speed (signal) X2EST (n) based on the equation (12). On the other hand, steps 570 to 599 of FIG. 13 and FIG.
It includes the portion of the provisional calculation of the routine of 60B (FIG. 7).

Step 500 to step 520 in FIG.
The operation of is described in steps 300 to 32 of FIG.
It is the same as the case of 0. Therefore, in order to simplify the description of the series of operations, the description of steps 500 to 520 is omitted here. Here, the formula (1
2) can be rewritten as the following [Equation 13].

[0112]

[Equation 13]

Furthermore, it will be readily appreciated that in step 525 of FIG. 12, the value of a is calculated by subtracting the sample at the immediately preceding position from the sample at the current position. Further, it will be readily appreciated that in step 530, the value of m is calculated by a multiplication operation (A * a = m). further,
In step 535 of FIG. 13, B * (K1.
By previously calculating U (n-1) + K2 · U (n-2)) = p, the above value of p may be calculated.
It will be easily recognized. Furthermore, it will be readily appreciated that in step 540, the value of q above is calculated by a multiply operation (E * X2EST (n-1) = q). Further, in step 545, m, p
By adding the values of q and q, the above X2EST
It will also be readily appreciated that the value of (n) is calculated.

The operations of steps 565 to 599 of FIGS. 13 and 14 are the same as those of steps 365 to 356 of FIG.
This is the same as in step 399. Therefore, here, in order to simplify the description of the series of operations, step 5
The description of 65 to step 599 will be omitted. FIG. 15 is a block diagram showing another embodiment of the velocity / position estimator used in the control device of the present invention. Here, a different type of velocity / position estimator 160A than the velocity / position estimator of FIGS. 6 and 7 is used.

Here, the velocity / position estimator 160 of FIG.
The operation of A is represented by the following equations (13) and (14).

[0116]

[Equation 14]

In these equations (13) and (14), the constant G2 represents the gain of the integrator, and has a value of 0.215, for example. The speed / position estimator 160A of FIG. 15 includes multipliers 251, 276 and a switch 25.
3 and 277, an integrator 290, and a delay circuit 292, which are different from the velocity / position estimator 160 of FIG.

The switches 253 and 277 are switched by a control algorithm executed in the determination circuit 155 of FIG. These switches 253, 27
When only the rough position information is used by the switching operation of No. 7, the constants C0 and C1 of the multipliers 250 and 275 are used.
Are each selected to provide a filter bandwidth of approximately 300 Hz. On the other hand, when only the precise position information is used, the constants C0 'and C1' of the multipliers 251 and 276 are used.
Are each selected to provide a filter bandwidth of approximately 1.5 kHz.

The integrator 290 calculates the position difference signal X1 (z).
-X1EST (z) is integrated and the estimated position signal X
1EST (z) and the estimated speed signal X2EST
Operates to present the integrated difference signal as used to obtain (z). The delay circuit 292 is
Operates to present the delayed and integrated differential signal.

The velocity / position estimator 160A of FIG. 15 and FIG.
Another difference from the velocity / position estimator 160 in FIG. 2 is that the delay circuit 225A has weighting constants K1 ', K2'. Here, K1 '= K1 / 0.5 = K
P · m, and K2 ′ = K2 / 0.5 = KP · (m−
1). Yet another difference between velocity / position estimator 160A and velocity / position estimator 160A is that multiplier 2
30A uses a constant D1 '(D1' = D1 · 0.5). The adder 295 operates to combine the delayed and integrated difference signal with the modified and past command signal. Furthermore, this adder 2
95 operates to supply the result thus combined to the multiplier 297. The multiplier 297 operates to multiply the signal supplied to this multiplier by 0.5.

Further, for the adder 245A, the integrator 2
An intermediate signal is generated by also adding the integrated difference signal as supplied from 90. The use of integrator 290 allows reducing the errors that can occur when the head approaches the target track and precise position information is used. Specifically, such an error may be caused by a drive bias force calibration error. If only rough position information is used, the benefit of integrator 290 is small.

Bias force calibration can be done in the factory or
This is performed immediately after the motor starts spinning up (Spin up). The calibration value thus obtained is stored in a table as a function of head position. However, prior to calibration, at least one erroneous calibration value has been stored in the table, resulting in inaccurate position and velocity estimates. In particular, the supplied positioner drive current must overcome the biasing force acting on the actuator, rather than producing acceleration. When this bias force is in the same direction as the acceleration, an acceleration having a value larger than the value corresponding to the command signal output is generated. On the other hand, when the bias force is in a direction other than the direction of acceleration, acceleration having a value smaller than the value corresponding to the command signal output is generated. In any case,
An error may occur.

FIG. 16 is a block diagram showing still another embodiment of the velocity / position estimator used in the control device of the present invention. The speed / position estimator 160C of FIG. 16 is generally the same as the speed / position estimator 160B of FIG. However, in the velocity / position estimator 160C of FIG.
An integrator is newly added. Then, the operation of the velocity / position estimator 160C in FIG. 16 is expressed by the following equations (15) and (16).

[0124]

[Equation 15]

[0125]

[Equation 16]

The constant G in equations (15) and (16)
3, G4 are adjusted by trial and error to obtain optimum performance. When making adjustments by trial and error, G
The first values of G3 and G4 are G3 = G2 · C0, G4 = 0.
It is set to 5.G2. Velocity / position estimator 16 of FIG.
0C includes a subtractor 475, an integrator 480, and a multiplier 48.
5, 495 and a delay circuit 490 are provided, which is different from the velocity / position estimator 160B in FIG.

The integrator 480 shown in FIG. 16 integrates the position difference signal X1 (z) -X1EST (z) to obtain the estimated position signal X1EST (z) and the estimated velocity signal X2EST (z). Present an integrated difference signal as used to acquire. Delay circuit 490 and multipliers 485, 495 have similar functions to the same components previously described.

FIGS. 17, 18 and 19 are diagrams for explaining the difference in effect between the speed / position estimator of FIG. 15 with and without the integrator. More specifically, FIG. 17 is a graph showing the transition of the drive current of the positioner when the seek operation is performed for 100 tracks in the speed / position estimator of FIG. 15, and FIG. 18 is the speed / position estimation of FIG. 19 is a graph showing a difference in speed estimation error (error in speed estimation) between the case where an integrator is provided and the case where no integrator is provided.
16 is a graph showing how a difference occurs in position estimation error (error when estimating the current position) in the velocity / position estimator of FIG. 15 with and without an integrator.

Here, the worst case is assumed in which the physical bias is applied to the simulation device of the disk device in the state where the correction of the output of the D / A converter is not performed. . Such a situation may occur when the disc drive has a 100% calibration error with respect to the worst-case expected bias force.

FIG. 17 shows the transition of the positioner drive current when the seek operation is performed for 100 tracks in the speed / position estimator of FIG. 15 when the maximum bias force is applied. . The curve of FIG. 17 shows the maximum acceleration. Furthermore, after some (2-3) sampling intervals, this maximum acceleration value changes to the maximum deceleration value. In FIG. 17, the operation of the velocity / position estimator using the precise position information is started at the boundary of the vertical line shown as F.
On the other hand, the disc device shifts from the seek mode to the track following mode with a vertical line indicated by M as a boundary. The track following position loop, which is separate from the control loop of FIG. 15, is outside the scope of the present invention and acts to lock the head in one position against biasing forces. Therefore, the steady state command currents cancel out to zero.

FIG. 18 illustrates how the velocity / position estimator of FIG. 15 has a difference in velocity estimation error with and without an integrator. The solid line in FIG. 18 shows the speed estimation error when the integrator 290 is not provided, that is, the operation of the speed / position estimator 160 in FIG. On the other hand, the broken line in FIG.
16 shows a velocity estimation error when using, that is, an operation of the velocity / position estimator 160A in FIG. The transient response in this case is slightly different depending on time, but is not a particular problem. In track following mode,
Do not use the speed estimate. Therefore, the state where the speed estimation error is zero on the rightmost side of FIG. 18 has no meaning.

The advantage of the integrator described above is that FIG. 18 and FIG.
It will be easily recognized by considering 9. 18 and 19, the velocity estimation error and the position estimation error are shown, respectively. The solid lines in these figures show the velocity estimation error and the position estimation error without the integrator 290. More specifically, when there is no integrator, in FIG. 18, there is a relatively large velocity estimation error even after the vertical line F, and in FIG. ) Exists.

Whatever the steady-state speed estimation error is, the head driving performance is settled (Settlin
g) has an adverse effect. This is because, if the mode is changed to the track following mode too early, an extra overshoot occurs due to the speed estimation error, resulting in an increase in seek time as a whole. On the other hand, if the transition speed to the track following mode is too slow, the phenomenon of head stall occurs and the seek time similarly increases.

The above position estimation error is only significant if erroneous position information is read from the disk based on either the Gray code or the position signal burst. In this case, an estimated value of the head position, that is, a position estimation error is used instead of the read position. When an error occurs in the control system due to the estimated value of the head position, the controller command is erroneously generated. The effect of the estimated value of the head position in this case is the same as that in the case of FIG. 6 described above.

By using an integrator, the steady-state errors mentioned above are essentially eliminated, and the consequences of these errors are avoided. Having thus far described specific embodiments of the invention, here
It is believed that it is merely illustrative of the present invention. Further, it is not desirable to limit the present invention only to the configurations shown in the present text, because those skilled in the art can easily make many variations and modifications. Accordingly, all suitable modifications and equivalents are conceivable as long as they come within the scope of the invention described in the claims appended hereto and their equivalents.

[0136]

As described above, according to the present invention,
During the seek operation, when the head moves toward the target track, only the rough position information contained in the servo information read from the intermediate track located between the current track and the target track is used. By estimating the current speed of the head in the disk device, the control current required for performing the seek operation of the head can be calculated by a simple calculation. As a result, while moving the head toward the target track quickly and quickly,
The head can be settled quickly during the track following operation.

[Brief description of drawings]

FIG. 1 is an enlarged view showing a part of a disc used in the present invention.

2 is a format diagram showing data items included in a servo frame in the disc of FIG. 1. FIG.

FIG. 3 is a diagram showing a speed control loop used in an apparatus for controlling a disk device of the present invention.

FIG. 4 is a graph showing the relationship between the control current of the voice coil motor generated in the control device of the present invention and the sampling time interval.

FIG. 5 is a graph showing the relationship between the head speed and the sampling time interval in the control device of the present invention.

6 is a block diagram illustrating one embodiment of a velocity / position estimator that may be used in the velocity control loop of FIG.

7 is a block diagram showing a modified example of the velocity / position estimator of FIG.

8 is a flowchart (No. 1) for explaining the operation of the speed / position estimator of FIG.

9 is a flowchart (part 2) for explaining the operation of the velocity / position estimator of FIG.

10 is a flowchart (part 3) for explaining the operation of the velocity / position estimator of FIG.

FIG. 11 is a flowchart (part 4) for explaining the operation of the velocity / position estimator of FIG. 6.

12 is a flowchart (part 1) for explaining the operation of the velocity / position estimator of FIG. 7. FIG.

13 is a flowchart (part 2) for explaining the operation of the speed / position estimator of FIG. 7. FIG.

14 is a flowchart (part 3) for explaining the operation of the velocity / position estimator of FIG. 7. FIG.

FIG. 15 is a block diagram showing another embodiment of the velocity / position estimator used in the control device of the present invention.

FIG. 16 is a block diagram showing still another embodiment of the velocity / position estimator used in the control device of the present invention.

17 is a graph showing changes in the drive current of the positioner when the seek operation is performed for 100 tracks in the speed / position estimator of FIG.

FIG. 18 is a graph showing how the speed / position estimator of FIG. 15 has a difference in speed estimation error with and without an integrator.

FIG. 19 is a graph showing a difference in position estimation error between the velocity / position estimator of FIG. 15 with and without an integrator.

[Explanation of symbols]

 105, 115 ... Subtractor 110 ... Velocity profile generator 120 ... Gain circuit 125, 145 ... Adder 130 ... Analog / digital (D / A) converter 132 ... Transconductance amplifier 135 ... Device set to be controlled 140 ... General position Demodulator 150 ... Precision position demodulator 155 ... Judgment circuit 160 ... Velocity / position estimator 180 ... Current sensor

Claims (26)

[Claims] 1. A seek of a movable head in a disk device of the type comprising a disk comprising a plurality of tracks having pre-written embedded servo information including rough position information and precise position information. A method for controlling motion, the method comprising: producing a position error indicative of a difference between a target track targeted by the head and a current track on which the head is located; and a velocity as a function of the position error. Presenting a command, generating a speed error indicating the difference between the speed corresponding to the speed command and the speed estimated for the head, and in response to the speed error Moving the head to the target track, the current track when the head moves toward the target track. Reading the servo information from an intermediate track located between the target track and the servo information read from the intermediate track without using the precise position information included in the servo information. A step of generating the estimated speed using rough position information. 2. The disk drive control method further determines a sample position based on rough position information included in the servo information, estimates an expected position of the head with respect to the disk, and arrives shortly thereafter. When the servo information is correct, the next current track is determined by using the sample position, while when the immediately following servo information is incorrect, the expected position is used to determine the next current track. The control method according to claim 1, further comprising the step of establishing a track. 3. The rough position information included in the servo information is used in the step of estimating the expected position without using the precise position information included in the servo information read from the intermediate track. The control method according to claim 2, which is used. 4. The step of reading the servo information from the disk is repeated by reading the embedded servo information from the disk at a predetermined sampling interval as the head moves over the disk. Each sampling interval consists of an in-line portion (In-line Portion) and a preliminary calculation (Precalculati
on Portion), the step of generating the estimated speed occurs during the period of the inline portion, wherein the expected position is the period of the provisional calculation portion. The control method according to claim 2, which is quoted therein. 5. The method of controlling the disk device further comprises the step of comparing the position of the next current track with the expected position to generate a feedback correction signal, The control method of claim 2, wherein the step of producing a determined speed is responsive to the feedback correction signal. 6. The method of controlling the disk device further comprises the step of generating an integrated correction signal by integrating the feedback correction signal, wherein the step of generating the estimated speed comprises: The control method according to claim 5, which is responsive to the integrated correction signal. 7. The method of controlling the disk device further comprises: if the current position of the head is within about 20 tracks from the target track, estimating the speed of the head at the current position by using the intermediate The control method according to claim 1, further comprising the step of using precise position information included in servo information read from the track. 8. The method of generating the estimated velocity operates within a predetermined bandwidth, wherein the method of controlling the disk drive further comprises: 20
8. The control method according to claim 7, further comprising increasing the predetermined bandwidth when the vehicle is within a track. 9. The disk drive control method further generates a control command in response to the speed error, and stores at least one value before the control command as previous control command data to be held. Further, at least one value before the estimated speed is stored as speed data estimated to be retained, and at least one value before the current track position is retained. Storing as previous track data, the step of moving the head responsive to the control command, and the step of generating the estimated speed including the previous control command to hold. The control method of claim 1, responsive to data, said estimated speed data to retain, and said previous track data to retain. 10. The step of generating the estimated speed comprises the step of comparing the previous track data to be retained with the position data of the current track to generate track difference data, A process of generating weighted track difference data by multiplying the track difference data by a predetermined value, the weighted track difference data, the previous control command data to be held, and the 10. The control method according to claim 9, further comprising the step of obtaining the estimated speed by combining the estimated speed data to hold the estimated speed. 11. The method of controlling a disk drive further comprises the step of estimating an expected position of the head as the head moves over the disk, the step of generating the estimated velocity. Comparing the previous track data to be held with the position data of the current track to generate track difference data, and comparing the position data of the current track with the expected position. To generate the position difference data, and to generate feedback correction data by multiplying the position difference data by a first value, the track difference data, the feedback correction data, and Combining the previous control command data to be held to generate an intermediate signal, and multiplying the intermediate signal by a second value, Generating the estimated intermediate signal and combining the weighted intermediate signal with the estimated velocity data to be retained to obtain the estimated velocity. 10. The control method according to claim 9, comprising. 12. The method of controlling the disk device further includes: the previous track data to be retained, the previous control command data to be retained, and the speed data estimated to be retained. 10. The control method according to claim 9, further comprising the step of estimating the expected position of the head as the head moves over the disk by coupling 13. Each seek operation includes an acceleration phase and a deceleration phase, the step of moving the head presenting a current command as a function of the speed error, and in response to the current command, Supplying the actual current to the motor of the disk drive, the motor being used to move the head, and the step of producing the estimated speed during the acceleration phase. Is the actual current value above, and
The control method according to claim 1, wherein a value corresponding to the current command is used during the deceleration stage. 14. A seek for a movable head in a disk drive of the type comprising a disk comprising a plurality of tracks having pre-written embedded servo information including rough position information and precise position information. A device for controlling motion, the means for generating a position error indicating a difference between a target track targeted by the head and a current track on which the head is positioned, and a speed as a function of the position error. Means for presenting a command, means for generating a speed error indicating a difference between the speed corresponding to the speed command and the speed estimated for the head, and, in response to the speed error, the current track from the current track. Means for moving the head to the target track; and, when the head moves toward the target track, the current track and the target track. Means for reading the servo information from an intermediate track located between the intermediate track and the track, and the rough position included in the servo information without using the precise position information included in the servo information read from the intermediate track. And a means for generating the estimated speed by using position information. 15. The control device of the disk device further determines a sample position based on rough position information included in the servo information, estimates an expected position of the head with respect to the disk, and arrives shortly thereafter. When the servo information is correct, the next current track is determined by using the sample position, while when the immediately following servo information is incorrect, the expected position is used to determine the next current track. 15. The control device according to claim 14, further comprising means for determining a track. 16. The rough position information included in the servo information is used by the means for estimating the expected position without using the precise position information included in the servo information read from the intermediate track. Claim 15 to respond
The control device described. 17. A means for reading the servo information from the disk reads the servo information at a predetermined sampling interval when the head moves on the disk to read the embedded servo information from the disk. Operative to read repeatedly, each of the sampling intervals including an in-line portion and a provisional calculation portion, wherein the means for generating the estimated speed operates during the in-line portion. 16. The control device according to claim 15, wherein the means for estimating the expected position operates during the part of the provisional calculation. 18. The controller of the disk device further comprises means for comparing the position of the next current track with the predicted position to generate a feedback correction signal, 17. The controller of claim 16, wherein the means for producing a determined speed is responsive to the feedback correction signal. 19. The control device of the disk device further comprises means for generating an integrated correction signal by integrating the feedback correction signal, wherein the means for generating the estimated speed comprises: The controller of claim 18, responsive to the integrated correction signal. 20. The control device of the disk device further comprises: if the current position of the head is within about 20 tracks from the target track, the intermediate position of the intermediate position in order to estimate the speed of the head at the current position. 15. The control device according to claim 14, further comprising means for using precise position information included in servo information read from the track. 21. The means for generating the estimated velocity operates within a predetermined bandwidth, the controller of the disk drive further comprising: a current position of the head approximately from the target track. 20
21. The control device of claim 20, comprising means for increasing the predetermined bandwidth when in track. 22. The control device of the disk device further includes means for generating a control command in response to the speed error, and control command data before holding at least one value before the control command. And further store at least one previous value of the estimated speed as estimated speed data to retain, and at least one previous value of the current track position, Storing means for storing as previous track data to be held, the means for moving the head is responsive to the control command, and means for generating the estimated speed is the holding 15. The controller of claim 14 responsive to previous control command data to be held, said estimated speed data to hold, and said previous track data to hold. 23. The means for generating the estimated speed compares the previous track data to be held with the position data of the current track to generate track difference data. Means for generating weighted track difference data by multiplying the track difference data by a predetermined value; the weighted track difference data; the previous control command data to be held; 23. Means for obtaining the estimated speed by combining the estimated speed data to hold the estimated speed. 24. The controller of the disk drive further comprises means for estimating an expected position of the head as the head moves over the disk, and means for generating the estimated velocity. Comparing the previous track data to be held with the position data of the current track to generate track difference data, and comparing the position data of the current track with the expected position. And means for generating position difference data, means for generating feedback correction data by multiplying the position difference data by a first value, the track difference data, the feedback correction data, and Means for generating an intermediate signal by combining previous control command data to be held, and weighting by multiplying the intermediate signal by a second value Means for producing the estimated intermediate signal and means for obtaining the estimated velocity by combining the weighted intermediate signal with the estimated velocity data to be retained. Item 23. The control device according to Item 22. 25. The control device of the disk drive further comprises: the previous track data to be held, the previous control command data to be held, and the speed data estimated to be held. 23. The control device according to claim 22, further comprising means for estimating an expected position of the head when the head moves on the disk by coupling the above. 26. Each seek operation includes an acceleration phase and a deceleration phase, the means for moving the head presenting a current command as a function of the speed error, and in response to the current command, the means for moving the head. Means for supplying the actual current to the motor of the disk drive, said motor being used for moving said head, said means for producing said estimated speed being provided during said acceleration phase. 15. The control device according to claim 14, wherein the actual current value is used and a value corresponding to the current command is used during the deceleration stage.