JPH087957B2 - Magnetic head digital positioning controller - Google Patents

Description

The present invention relates to a digital positioning control device for a magnetic head of an FDD device (floppy disk device) by a sector servo system.

[Prior art]

In the FDD device, a step motor is used as an actuator for positioning the magnetic head because of the simplification of the servo mechanism and the size and cost of the device.

However, the eccentricity of the data tracks cannot be ignored as the pitch between the data tracks of the magnetic disk becomes narrower as the size and capacity increase, so the step motor is feedback-controlled by the sector servo method instead of the open-loop control, and positioning is performed. The accuracy is improved.

As is well known as the sector servo system, the data surface of the magnetic disk is divided into several sectors, and the track position information is written at the head of this sector, so that the track position information is obtained at each sampling time determined by the number of sectors and the rotation speed of the magnetic disk. In this method, feedback control is performed.

Conventionally, in the feedback control by the sector servo system in the FDD device, as described in Japanese Patent Application Laid-Open No. 58-151613, the magnetic head is stepped by a step motor at a rate of one step per sector according to track position information obtained from the beginning of the sector. According to the method, the method of feeding inward or outward is adopted.

[Problems to be solved by the invention]

In this method, the data track follow-up error of the magnetic head becomes approximately equal to the feed pitch of the magnetic head determined by the maximum eccentricity amount per sector of the data track, so that the data track density is made higher unless the number of sectors is increased. There is a problem that it is difficult. Further, since the dynamic characteristic of the step motor has the characteristic of the spring-mass system, the magnetic head becomes oscillating in the positioning by the step feed, which makes it difficult to perform highly accurate positioning.

The present invention was devised for the purpose of compensating for the vibration characteristics of the step motor and further improving the follow-up characteristics of the magnetic head to the eccentric data track.The feed pitch is determined by the mechanical structure of the step motor. Nevertheless, it is an object of the present invention to provide a digital positioning control device that positions a magnetic head on a data track with high accuracy by minutely changing the position of a step motor (that is, the position of a magnetic head).

[Means for solving problems]

In order to achieve the above object, the present invention is provided with a floppy disc drive including position error detection means for obtaining a position error signal indicating a difference between a target position to be followed by the magnetic head and the position of the magnetic head at every constant sampling time. In a digital positioning control device for a magnetic head in a disk device, a step motor for driving the magnetic head, an integrating means for integrating the position error signal at each sampling time, and an output of the integrating means for receiving the target. The digital filter for generating a sine wave having the same frequency as the second frequency component in the position fluctuation component with respect to the impulse input, the output of the digital filter, the position error signal, and the output of the integrating means are sampled. The stabilized digital filter that is input at each time and the stabilized digital filter A storage element that outputs an exciting current value to each phase of the step motor according to the output of the step motor, and an amplifier that applies a current to the step motor through a low-pass filter after a fixed time of the sampling time according to the output value of the storage element. And a minute feed circuit of a step motor including the.

[Action]

In the FDD device, a step motor is used as an actuator for positioning the magnetic head from the viewpoint of cost and size, but it is necessary to position the step motor with higher accuracy in order to reduce the size and increase the capacity. . The minute feed circuit stores the relationship between the exciting current to each phase of the step motor and the equilibrium position,
This enables minute positioning of the step motor regardless of the feed pitch determined by the mechanical structure of the step motor.

Also, in the sector servo system in the FDD device, the number of sectors and the number of rotations of the magnetic disk cannot be increased so much,
Therefore, the sampling frequency does not necessarily become high. In order to make the magnetic head follow the eccentric data track under such an environment, the second component sine wave that generates a sine wave having a frequency equal to twice the rotation frequency of the magnetic disk, which is the dominant mode of eccentricity, is generated. Digital filter 1
By inserting in a circular loop in series, the steady tracking characteristic of the magnetic head is improved, and as a result, it is possible to reduce the position error between the magnetic head and the data track position.

〔Example〕

Hereinafter, the present invention will be described in detail with reference to the drawings. However, in the following description, signal names are represented by the same symbols for signal values.

FIG. 1 is a block diagram showing the configuration of a digital positioning control device for a magnetic head using a two-phase linear step motor according to an embodiment of the present invention. Assuming that the sampling time is T, a position error a indicating the difference between the target position to be followed by the magnetic head 6 and the position x of the magnetic head (not explained here, but obtained by some position error detecting means).
Is input to the integrator 1 and the stabilizing digital filter 3 every sampling time T.

The integrator 1 integrates the position error a for each sampling time, and outputs the integrated value b to the second component sine wave digital filter 2 and the stabilizing digital filter 3 during the sampling time T.

The second component sine wave digital filter 2 is a magnetic head 6
Is a digital filter for generating a sine wave having a frequency equal to the second frequency component in the fluctuation component of the target value signal to be followed, to the impulse input.

The stabilizing digital filter 3 includes a position error signal a, an output b of the integrator 1, and a second component sine wave digital filter 2
The output c of the two-phase linear step motor 5 is calculated, and the minute feed circuit 4 calculates the balance point address value u of the sampling time T, that is, at (T + Δ) time.
Two-phase linear step motor 5 during sampling time T
And outputs the equilibrium point address value u. Δ is a delay time corresponding to the calculation time of the stabilized digital filter.

In the floppy disk device, a magnetic disk on which data is recorded uses a plastic film such as polyethylene as a base on which a magnetic medium is applied. Therefore, the data tracks on which the data is recorded are formed on concentric circles, but the plastic film, which is the base of the magnetic disk, expands and contracts depending on the environment in which the data is recorded (temperature, humidity, etc.), and the data tracks are deformed. It is known that the deformation of this data track is elliptical. As a result, the target data track position to be followed by the magnetic head 6 fluctuates at a frequency twice the magnetic disk rotation frequency. The second component sine wave digital filter has the effect of increasing the feedback gain at this frequency in the loop transfer function of the positioning digital control system of the magnetic head.
As a result, the track following accuracy of the magnetic head can be improved.

The integrator 1 and the second component sine wave digital filter 2 are steady-state deviation compensators generally called internal models, and the stabilizing digital filter 3 is a compensator for stabilizing the feedback control system.

The minute feed circuit 4 receives the output value u of the stabilizing digital filter 3 and supplies it to each phase (A phase and B phase) of the two-phase linear step motor 5 according to the output value u of the stabilizing digital filter 3. , d ′ are caused to flow, the equilibrium point of the two-phase linear step motor 5 is changed, the two-phase linear step motor 5 is moved, and the magnetic head 6 is positioned on the data track. The position error a between the position x of the magnetic head 6 and the target track position to be followed by the magnetic head 6 is detected again at the next sampling time and is input to the integrator 1 and the stabilizing digital filter 3, whereby the feedback loop is formed. Composed. Here, the sampling time T and the delay time Δ are managed by a timer.

FIG. 2 is a block diagram showing one configuration example of the minute feed circuit 4. The minute feed circuit 4 receives the equilibrium point address value u of the two-phase linear step motor 5 given from the outside, and
ROM (read-only memory) 7,7 'that outputs the exciting current value to be given to each of the A-phase and B-phase of the single-phase linear step motor and the digital signal that is the output value e, e' of ROM 7,7 'is an analog signal D / A converters 8 and 8'to convert to D / A
Nyquist frequency (1/2 of output signal f, f 'of converter 8,8'
Low-pass filter 9,9 'that cuts off frequency components above T)
And the currents d and d'in the A-phase and B-phase of the two-phase linear step motor 5 according to the outputs g, g'of the low-pass filters 9, 9 '.
It is composed of amplifiers 10 and 10 'which flow.

The 2-phase linear step motor 5 moves the position of its equilibrium point
By properly combining the A-phase current value d and the B-phase current value d ', it can be arbitrarily set regardless of the mechanical pitch of the two-phase linear step motor 5. That is, when a certain amount of current is applied to each of the A phase and the B phase, there is always an equilibrium point due to the combination of the current values, and the position of this equilibrium point depends only on the excitation currents of the A phase and the B phase. It has nothing to do with the feed pitch determined by the mechanical structure.

Therefore, when the equilibrium point position due to a certain combination of currents is used as the reference address (address value 0), the amount of deviation from that point can be used as the equilibrium point address of the linear step motor. , A combination of current values corresponding to the equilibrium point address is stored. However, the combination of the A-phase current value and the B-phase current value is taken so that the maximum static thrust becomes constant. For example, when the tooth pitch of the two-phase linear step motor is λ, A
Assuming that the equilibrium point position of the two-phase linear step motor 5 when a current of +0.1 amp is applied to both phase B and phase B is the reference address, and the maximum stationary thrust at this time is the reference maximum stationary thrust, the equilibrium point Position is λ / 8 from the reference address
Up to +0.1 amp (address 0) for ROM7
To 0.0 amps (address λ / 8) to phase A, and ROM7 'from +0.1 amps (address 0) The value for flowing the current up to (address λ / 8) in the B phase is stored in sequence according to the address of the equilibrium point. FIG. 3 shows this state. The reference address (address value 0) determined by a certain combination of currents and the equilibrium point address u, which is the deviation from that point,
The static thrust characteristics at u'u "are shown.
The position x of the phase linear step motor 5 (magnetic head 6) is also given as a position deviation from the reference address.

FIG. 4 is a circuit diagram showing a specific configuration example of the low-pass filters 9, 9 '. The low-pass filter of FIG. 4 is represented by the quadratic transfer function Glp (s) shown below.

However, s is a Laplace transform operator.

There is the following relationship between the equation (1) and the circuit constant in FIG.

As described above, the minute feed circuit can arbitrarily select the position of the equilibrium point of the two-phase linear step motor 5 regardless of the feed pitch determined by the mechanical structure.

When the fine feed circuit 4 is connected to the 2-phase linear step motor 5, as shown in FIG. 3, the equilibrium point address value u of the 2-phase linear step motor 5 and the 2-phase linear step motor 5 (magnetic head 6 The input-output relationship between the position x and the position x) is expressed by equation (4).

M + kx = ku (4) Here, M is the mass of the movable part driven by the two-phase linear step motor 5, k is a constant determined by the feed pitch λ of the two-phase linear step motor 5 and the maximum static thrust F, It is expressed by equation (5). Further, is the second derivative of x with respect to time, that is, the acceleration α.

In a system in which the equilibrium point address value u of the two-phase linear step motor 5 is input and the position x of the magnetic head 6 is output, as shown in FIG.
A dead time corresponding to the time calculation delay is added, and t =
Paying attention to the sampling time of NT (N = 0,1,2, ...) and deriving the equation of state at discrete time values, it becomes as shown in equation (6). In FIG. 5, v is an input to the dead time 11.

X ((N + 1) T) = AX (NT) + bv (NT) (6-
a) Y (NT) = C * X (NT) (6-b) where X (NT) is a state vector and Y (NT) is an output vector, which is given by the following equation.

X (NT) = [x (NT), β (NT), u (NT)] ^T Y (NT) = [x (NT), u (NT)] ^T where β is the speed of the magnetic head 6. Represent.

Further, each coefficient matrix of the equation (6) has the following structure.

However, exp (•) is a matrix exponential function.

The system represented by equation (6) is controllable and observable, and the observable index is 2.

FIG. 6 is a block diagram showing the function of the accumulator by using the Z-transform operator z. z ^-1 is a shift register that operates at each sampling time T, integrates the position error signal a at each sampling time, and outputs an integration result b at each sampling time T for the sampling time.

FIG. 7 is a block diagram showing a configuration example of the second component sine wave digital filter 2 described using the Z-transform operator z. In FIG. 7, m1, m2, m3, m4, m5, m6 are real constants. C1 and C2 represent the outputs of the second component sine wave digital filter. Since the magnetic disk sheet in the FDD device deforms into an elliptical shape due to changes in temperature and humidity, the eccentric mode of the data track appears at a frequency about twice the rotation frequency of the magnetic disk. That is, the follow-up control system of the magnetic head 6 has about 2 times the rotation frequency of the magnetic disk.
It must have sufficient tracking accuracy for the target value signal that fluctuates at twice the frequency. Widening the bandwidth of the control system is considered as a method for this, but FDD
In the sector servo system in the device, the number of sectors and the number of rotations of the magnetic disk cannot be set so large that the sampling frequency does not become so large and the bandwidth of the control system cannot be widened. Second
The component sine wave digital filter 2 is a digital filter added for the purpose of causing the magnetic head 6 to follow the eccentricity of the data track with sufficient accuracy even when the sampling frequency is low.

In FIG. 7, the pulse transfer function Gsin from the accumulator output b to the output C1 of the second component sine wave digital filter 2
(Z) is expressed by equation (7).

For example, when the eccentricity mode of the data track is 10 Hz and the sampling time T is 6 msec, Laplace transformation of By discretizing at sampling time T, m1, m
2, m3, m4, m5, m6 are as follows. However, t represents time.

m1 ＝ 1.7788 × 10 ^-5 , m2 ＝ 5.8590 × 10 ^-3 m3 ＝ 5.8590 × 10 ^-3 , m4 ＝ 0.92985, m5 ＝ 0.92985, m6 ＝ －23.10706 Is stabilized. At this time, as for the pulse transfer function W (z) from the target position to be followed by the magnetic head 6 to the position error a, the one-cycle pulse transfer function is the equation (7) and the pulse transfer function 1 / (z-1) of the integrator. Because it contains
It always looks like this:

Here, δ (z) is a σ-order real coefficient polynomial in which all roots of δ (z) = 0 exist in the unit circle of the Z-plane, and γ
(Z) is a real coefficient polynomial of degree (σ-3). Also, σ
Is a natural number.

The pulse transfer function (frequency characteristic) from the target position to be followed by the magnetic head 6 represented by the equation (8) to the position error a (t) is as shown in FIG.

FIG. 9 shows that the position error a (t) is greatly compressed in the DC region with a frequency twice the rotation frequency of the magnetic disk.

This can be shown quantitatively using the Z-transform final value theorem.

Since the formula (8), the theorem of the final value of the Z-transform, and the eccentricity of the data track occur at a frequency twice the rotation frequency of the magnetic disk, the position error a (t) is = 0 (9) a (t) means a position error between the magnetic head 6 and the data track position to be followed by the magnetic head 6. Therefore, when the target position is eccentric at a frequency twice the rotation frequency of the magnetic disk, the position error a (t) is expressed by the equation (9).
Shows that it constantly becomes zero, supporting the frequency characteristic shown in FIG. That is, due to the effect of the second component sine wave digital filter, the magnetic head 6 can accurately follow the frequency twice the rotational frequency of the magnetic disk, which is the main component of the fluctuation of the target position. Further, the position error up to the control band is compressed by the integrator 1 along the asymptote having a rate of 20 dB per decade as shown in FIG.

FIG. 8 shows an example of the structure of the stabilizing digital filter. As mentioned above, the system expressed by equation (6) is controllable.
Since it is observable and the observable index is 2, the control system can be stabilized by a first-order stabilizing digital filter.
Therefore, in this case, the order σ of σ (z) in the expression (8) is 7. In FIG. 8, l1, l2, l3, l4, l5, l6, l7, l8, l
9, l10 are real constants. By selecting these properly, the denominator polynomial δ (z) in equation (8) can be specified arbitrarily, and although there is a limit due to the sampling frequency, control is performed within that range. The band width of the system can be set arbitrarily.

The second component sine wave digital filter and stabilizing digital filter shown in FIGS. 7 and 8 are shift registers, adders, m1, m2, m3, m4, m5, m6 or l1, l2, l3, l4, l.
It can be constituted by a multiplier for multiplying 5, l6, l7, l8, l9, l10 by a constant and a timer for controlling the sampling time T and the timing Δ for outputting the equilibrium point address value u to the two-phase linear step motor 5. Needless to say, the real constants m1, m2, m
3, m4, m5, m6, l1, l2, l3, l4, l5, l6, l7, l8, l9, l10 are real constants generally including zero. Therefore, in the case of 0, among the components shown in FIG. 7 and FIG. 8, those that are multiplied by 0 do not have to be components and can be omitted.

As a matter of course, the integrator 1, the second component sine wave digital filter 2 and the stabilizing digital filter 3 are
Since it is represented by the Z-transform operator z as shown in FIGS. 6, 7, and 8, it is possible to programmatically solve the difference equation by a digital computer such as a microprocessor. It can also be realized.
Furthermore, the present invention is not limited to the case of using a two-phase linear step motor, but can be applied to an FDD device using another step motor.

〔The invention's effect〕

As described above, according to the present invention, by adding the minute feed circuit to the step motor and further inserting the second component sine wave digital filter in series in the loop loop, the feed determined by the mechanical structure of the step motor is obtained. The step motor can be fed finely regardless of the pitch, and the magnetic head can follow the target data track with high accuracy even under a low sampling frequency.

[Brief description of drawings]

FIG. 1 is a block diagram showing the construction of a digital positioning control device for a magnetic head using a two-phase linear step motor according to an embodiment of the present invention, and FIG. 2 is a construction of the minute feed circuit shown in FIG. FIG. 3 is a block diagram showing an example, FIG. 3 is a diagram showing addresses of respective equilibrium points of a two-phase linear step motor and static thrust characteristics of the two-phase linear step motor at that point, and FIG. 4 is a low pass shown in FIG. FIG. 5 is a block diagram showing a specific configuration of a filter, FIG. 5 is a block diagram of a control target with dead time added, FIG. 6 is a block diagram showing a function of an integrator, and FIG. 7 is a second component sine wave digital filter. Is a block diagram describing one configuration example of the Z-transform operator z, FIG. 8 is a block diagram illustrating one configuration example of the stabilized digital filter using the Z-transform operator z, and FIG. Magnetic head Position error a (t) from the target position that 6 should follow
It is the figure which showed the pulse transfer function up to. 1 …… Integrator 2 …… Second component sine wave digital filter 3 …… Stabilizing digital filter 4 …… Fine feed circuit 5 …… 2-phase linear step motor 6 …… Magnetic head 7,7 ′ …… ROM 8, 8 '... D / A converter 9,9' ... low-pass filter 10,10 '... amplifier 11 ... dead time a ... position error signal b ... integrator output c, c1, c2 ... second Component sine wave digital filter output d, d '... minute feed circuit output e, e' ... ROM output f, f '... D / A converter output g, g' ... low-pass filter output u, u ', u ″ …… Balance point position of 2-phase linear step motor v …… Input to dead time x …… Magnetic head position R ₁ , R ₂ …… Resistance C ₁ , C ₂ …… Capacitors l1, l2, l3, l4, l5, l6, l7, l8, l9, l10, m1, m2, m3, m4, m5, m6…
2 times the frequency of the ... real constant z ...... Z- conversion operator ωb ...... control band frequency ω ₂ r ...... disk rotation frequency

Claims (1)

[Claims] 1. A magnetic head digital in a floppy disk device comprising position error detection means for obtaining a position error signal indicating a difference between a target position to be followed by the magnetic head and the position of the magnetic head at every constant sampling time. In the positioning control device, a step motor for driving the magnetic head, an integrating means for integrating the position error signal for each sampling time, and an output of the integrating means to receive a change component of the target position. , Second
A digital filter for generating a sine wave having the same frequency as a frequency component with respect to an impulse input, and a stabilizing digital filter having the output of the digital filter, the position error signal and the output of the integrating means as input at each sampling time. A storage element that outputs an exciting current value to each phase of the step motor according to the output of the stabilizing digital filter, and the step motor to the step motor according to the output value of the storage element via a low-pass filter after a fixed time of the sampling time. And a minute feed circuit for a step motor including an amplifier for applying a current to the magnetic head.