JPH03160674A - Offset removal system for disk position signal - Google Patents

Abstract

PURPOSE:To suppress a track pitch error to be a minimum and to improve the control precision of a tracking servo by detecting the offset quantity of a position signal, which is obtained by actually moving a head and removing the offset quantity from the position signal. CONSTITUTION:A cross time interval detection part 18 detects the time interval of an upper side and that of a lower side in the reference level of the position signal when the servo head 12 is moved in the direction of the diameter of a disk. The time intervals are given to an offset time interval detection part 20 and an offset time interval is detected as a time deviation quantity as against the position signal without an offset in the direction of a time base. Then, an inclination detection part 22 calculates the inclincation of the signal from the upper and lower time intervals and the amplitude value of the position signal. Then, the offset time interval and the inclination are multiplied so as to calculate the offset quantity, and the offset quantity is removed from the subsequent position signal. Thus, the track pitch error is set to be the minimum and the control precision of the servo can be improved.

Description

[Detailed Description of the Invention] [Summary] This invention relates to a disk position signal offset removal method that automatically removes the offset of a position signal created from a servo pattern read signal, and improves control accuracy and recording density by removing the position signal offset. When setting the calibration mode, the clock time interval above and below the reference level of the position signal obtained when the servo head is moved at a constant speed in the radial direction is detected. It is configured to obtain the offset time interval and slope, calculate the offset amount by multiplying the offset time interval by the slope, and store it in memory, and then remove the stored offset amount from the position signal obtained and output it. do.

[Industrial Application Field] The present invention relates to a disk position signal offset removal method that automatically removes the offset of a position signal created from a servo pattern read signal.

In a magnetic disk device, when receiving an access command, a seek operation (coarse control) is performed to move the head to an access track, and when the seek operation is completed, an on-track operation (fine control) is performed to cause the head to follow the track.

The seek operation and on-track operation of the head in such a magnetic disk device are performed based on a position signal generated from a read signal of a disk medium on which a servo pattern is recorded. That is, a servo pattern for creating a position signal indicating the radial position of the head is recorded on one of a plurality of disk media, this servo pattern is read by a dedicated servo head, and the read servo signal is read. A position signal based on the peak value, e.g. a two-phase position signal POS
N. Creating POSQ.

However, the position signal also has an offset due to the offset of the peak detector and differential amplifier installed in the servo demodulation circuit that creates the offset signal, which appears as a track pitch error. A method to remove it is desired.

[Prior Art] Conventionally, in a magnetic disk device equipped with a servo disk having a servo pattern recorded thereon and a servo head, a servo signal read by the servo head is input to a servo demodulation circuit to generate, for example, a two-phase position signal POSN. , is creating POSQ.

That is, the peak values of four amplitude changes that occur during the synchronization pattern of the servo signal are detected by a peak detector to produce peak outputs A, B, C, and D, and the peak outputs A, B and the reference level R are differentially output. Input to amplifier and position signal POS
Find N as POSN=A-B+R, and similarly calculate peak outputs D, C and reference level R.
is input to another differential amplifier to obtain the position signal POSQ as POSQ=C-D+R.

In this case, if there is no offset in the peak detector or differential amplifier of the servo demodulation circuit, the two-phase position signals POSN and POSQ, which change symmetrically up and down with reference level R as shown in Figure 10(a), will be In this case, the track pitch Tp determined by the cross interval T between the reference level R of the position signals POSN and POSQ is always constant.

[Problems to be Solved by the Invention] However, an offset occurs in the position signal due to the offset of the peak detector and differential amplifier provided in the servo demodulation circuit.

For example, as shown in FIG. 10(b), the position signals POSN and POSQ have different offsets that are asymmetrical above and below the reference level R, and this offset appears as a track pitch error.

For this reason, if an offset occurs in the position signal, the precision of track servo control is reduced, and it becomes an obstacle to increasing the density of tracks, which in turn limits the expansion of disk storage capacity.

The present invention has been made in view of these conventional problems, and it is an object of the present invention to provide a disk position signal offset removal method that can improve control accuracy and recording density by removing position signal offsets. .

[Means to solve the problem] FIG. 1 is a diagram explaining the principle of the present invention.

First, the present invention has the same amplitude pattern repeated in the circumferential direction,
A disk medium 10 in which a servo pattern in which different amplitude patterns of different types are repeated in the radial direction in units of multiple tracks is recorded, a servo head 12 that reads the servo pattern recorded on the disk medium 10 and outputs a servo signal, and a servo head 12 that reads the servo pattern recorded on the disk medium 10 and outputs a servo signal. servo demodulation means 14 for creating a position signal indicating the radial head position with respect to the disk medium 10 based on the amplitude peak value of the servo signal from the head 12; A head control means 1 that performs a seek operation to move the head to a position and an on-track operation to cause the head to follow a track.
The target is a servo control circuit for a disk device equipped with 6.

In the disk position signal offset removal method of the present invention for such a servo control circuit of a disk device,
When the servo head 12 is driven at a constant speed in the disk radial direction, the position signal obtained from the servo demodulation means 14 is at the reference level R.
A first time interval T1 from when the position signal crosses the reference level R in the upward direction until the next time it crosses it in the downward direction, and a second time interval T2 from when the position signal crosses the reference level R in the downward direction until it crosses it in the next upward direction. a cross time interval detection means 18 for detecting; an offset time interval detection means 20 for detecting an offset time interval Tel indicating a shift in the time axis direction of the position signal based on the first and second time intervals T1 and T2;
a slope calculation means 22 that calculates the slope K of the position signal based on the offset time interval TO1 and the prescribed amplitude value V of the position signal; An offset calculation means 24 for calculating and storing Vof is provided, and the offset amount Vo1 is removed from the position signal from the servo demodulation means 14 and outputted to the head control means 16.

Here, the cross time interval detection means 16 detects a plurality of first time intervals Tll, TI2, ... Tln and second time intervals T21, T22, Tln from the position signal obtained from the head start to the stop. ... After detecting T2a, each average value is calculated.

Further, the offset time interval detection means 20 calculates the offset time interval Tof by dividing the difference between the first and second time intervals (Tl-72) by 4.

Tel= (T 1 +7 2) /4 Furthermore, the slope detection means 24 doubles the amplitude voltage V, 2■, to the first and second
The slope K is calculated by dividing by the sum of the time intervals (T1+72).

K=2V/ (T1+T2) [Function] According to the disk position signal offset removal method of the present invention having such a configuration, the offset amount of the position signal is automatically measured and calculated by setting the caliption mode. When the operation is performed and subsequent accesses are made, the offset is removed by adding or subtracting the stored offset amount to the position signal from the servo demodulation circuit, thereby minimizing the track pitch error. This makes it possible to improve control accuracy, increase track density, and greatly contribute to increasing storage capacity.

[Embodiment] FIG. 2 is a block diagram showing an embodiment of the present invention.

In FIG. 2, a servo head 12 reads a servo pattern stored on a servo disk (not shown) and outputs a read signal to the servo demodulation circuit 14. In this embodiment, a two-phase servo pattern shown in FIG. 4 is stored on the servo disk. That is, the magnetic recording state is shown by taking out five servo tracks on the left side of FIG.
Four types of servo patterns, 3 and P4, are recorded, and for the fifth servo pattern, the same servo pattern P1 as the first appears, and this is repeated in the radial direction thereafter. Naturally, the pattern on each track is recorded repeatedly in the circumferential direction.

In FIG. 4, the servo heads 12 are simultaneously located on the right side of the illustrated cylinders CYO, CYI. It shows the signal waveforms of the read signal and the servo signal obtained when positioned at CY2. For example, when the servo head 12 is positioned at cylinder CYO, the servo signal waveforms are A,
At four timings B, C, and D, for the P1 pattern, small amplitude patterns are obtained at A and B, no amplitude pattern at C, and a large amplitude pattern at the last D.

Referring again to FIG. 2, the servo signal from the servo head 12 is applied to a servo demodulation circuit 14. That is, the servo signal is first amplified by the AGC amplifier 44, and then amplified by the four peak detectors 46-1.46-2.46-3.46.
-4 performs peak detection. As shown in the timing chart of FIG. 5, the peak detectors 46-1 to 46-4 detect A during the synchronization pattern of the servo signal of FIG. 5(a) (the servo signal of cylinder CYO of FIG. 4). , B, C, D, and the current servo signal amplitude peak value exceeds the previous peak value held in the capacitor. and,
As shown in the figure, the process of charging the built-in capacitor to the peak level of the servo signal is sequentially repeated for each synchronous cycle. Note that after charging by the wind pulse, the capacitor charge is discharged slowly at a discharge current that is less than 1/10 of the charging current, so the peak does not peak with a gentle linear slope as shown in the figure until the next peak detection cycle. Level decreases.

Referring again to FIG. 2, peak detector 46-1.
The peak detection outputs A and B of 46-2 are the differential amplifier 48
-1 is input. A reference voltage R is applied from the reference voltage generator 50 to the differential amplifier 48-1, and the differential amplifier 48-1 generates a position signal POSN as follows: POSN=A-B+R. Also, peak detector 46-3.4
The peak detection outputs C and D of 6-4 are the differential amplifier 48-
2, and the differential amplifier 482 uses the reference voltage R from the reference voltage generator 50 to generate the position signal P.
Create OSQ as POSQ=C-D+R.

FIG. 6 shows a signal waveform diagram of the position signals POSN, POSQ (when the head is moved at a constant speed in the disk radial direction) output from the differential amplifier 48-1, 48-2 in FIG. With respect to the position signal POSN shown by the solid line, the position signal POSQ shown by the broken line has a phase delay of 1/4 cycle, that is, a phase difference of 1 track pitch.

Referring again to FIG. 2, each of the position signals POSN and POSQ produced by the servo demodulation circuit 14 is converted into a digital signal by the AD converter 34, and then taken into the MPU 36. The MPU 36 has a function as the head control means 16 shown in the principle explanatory diagram of FIG. 1, and also performs control processing for removing the offset of the present invention, which will be explained later.

When the MPU 36 receives read or write access from the host controller, it calculates a track difference as the difference between the current track position and the target track position, generates a target speed according to the track difference, and adjusts the head speed calculated from the position signal. The deviation between the actual speed and the target speed is determined, and this deviation data is converted into an analog deviation voltage by the DA converter 38 and supplied to the power amplifier 40, and the head motor 42 is driven to move the servo head 12 in the disk radial direction, that is, in the target track direction. Perform a seek operation to move to . When the seek operation by the MPU 36 is completed and the head reaches the target track position, the on-track operation is switched.

In this on-track operation, the MPU 36 calculates the deviation of the position signal with respect to the track center, and the position deviation data is sent to the DA.
The converter 38 converts the voltage into an analog deviation voltage, and the power amplifier 40 drives the head motor 42 to cause a series of heads including the servo head 12 to follow a target track (target cylinder position).

FIG. 3 is a functional block diagram of the offset removal method of the present invention implemented by the MPU 36 of FIG. 2.

In FIG. 3, the two-phase position signals POSN and POSQ from the servo demodulation circuit 14 are sent to adders 30 and 32, respectively.
It is provided to the servo control section 16 side for performing seek operations and on-track operations via changeover switches SW1 and SW2. When the calibration mode is set in the disk device, the changeover switches SW1 and SW2 are switched to the b side, and the two-phase position signals POSN and POSQ obtained via the adders 30 and 32 are transferred to the cross time interval detection section 18.
give to

When setting the calibration mode, the MPU 36 moves the servo head 12 from a predetermined start position in the disk radial direction at a certain moderate and low speed in order to actually measure the amount of offset of the position signal. The amount of offset is detected based on two-phase position signals POSN and POSQ obtained by moving the head for plation.

That is, the cross time interval detection unit 18 in FIG. 3 detects position signals POSN, P obtained when the servo head 12 is moved in the disk radial direction by setting the calibration mode
For each OSQ, as shown in FIG. Time interval T2 (second time interval) from crossing in the downward direction until crossing in the upward direction next time
Detect. Actually, as shown in FIG. 8, the average value of the time intervals of position signals of a plurality of cycles is calculated. The upper time interval T1 and the lower time interval T2 of the reference level detected by the cross time interval detection unit 18 are given to the offset time interval detection unit 20, and are applied to the offset time interval detection unit 20. The offset time interval Tel is detected as the amount of time shift of the position signal. Subsequently, a tilt detector 22 is provided, and the tilt detector 22 detects the first and second time intervals Tl obtained by the cross time interval detector 18.
, T2 and a predetermined amplitude value V of the position signal, the slope K of the position signal is calculated. Then, in the offset calculating section 24 provided finally, the offset time interval T determined by the offset time interval detecting section 20 is calculated.
The position signal offset ffi V o I is obtained by multiplying el by the slope K of the position signal obtained by the slope detection section 22.
and for the position signal POSN, Vat (N)
The position signal POSQ is set in the register 26 as Vof (Q).

When the offset amount is set in the registers 26 and 28 in this way, the calibration mode is canceled and the changeover switches SW1 and SW2 return to the a side again, and in this state, the two-phase position signal POSN obtained from the servo demodulation circuit 14 ,P
Registers 26.32 with adders 30.32 for each of the OSQs.
The position signal is corrected by the offset amount stored in 28, specifically, the offset amount is added or subtracted to remove the offset from the position signal, and the result is output to the servo control section 16.

Next, the offset removing operation of the position signal according to the present invention shown in FIG. 3 will be explained with reference to the operational flow diagram of FIG. 7.

First, when the calibration mode is set, the servo head is driven at a low and constant speed in a fixed direction, which is the disk radial direction, in step SL (hereinafter, "step" is omitted). Next, in S2, the PO obtained when moving the head
SN and POSQ are sampled via the AD converter 34. By sampling S2, for example, the 8th
The position signals POSN and POSQ shown in the figure are obtained.

Next, in step 83, the time TN(1) from when the position signal POSN crosses the reference voltage (reference level) R in the upward direction until it crosses the reference voltage R in the downward direction is measured. Next, the time TN (2) until the position signal crosses in the upward direction is measured. Similarly, TN (3), TN (4), ..., TN (100)
Measure up to.

Similarly, for the other position signal POSQ, time TQ(1) . TQ
(2) ,..., measure TQ(100).

Next, the process proceeds to S4, where the average value T (N+) of the upper time interval of the reference voltage R and the average value T (N-) of the lower time interval of the reference voltage R regarding the position signal POSN.
). Specifically, as shown in the figure, the average value T (N
10), the time interval TN (i
), the sum of the 50 measurement times for which i is an odd number is calculated and divided by 50. On the other hand, for the average value T (N-), T
The sum of 50 measurement times for which i of N(i) is an even number is calculated and divided by 50. As a result, as shown by the broken line in FIG. 9, the upper time interval T1=T(N+) of the reference level R in one cycle of the measurement position signal and the lower time interval T
2=T (N-) has been found.

Next, based on the average time interval between the upper and lower sides of the reference voltage R, the offset time interval Tel is calculated as the amount of deviation in the time axis direction of the position signal that actually has an offset compared to the case without an offset. As shown in the figure, this offset time interval Tof is the difference between the average values of the upper and lower time intervals of the reference voltage (T (N+) - T (N-) = 71
-72) by 4. Specifically, as shown in FIG. 9, for example, if TI=10 and T2=6, the offset time interval Tel in this case is Tol
= (T1-T2) /4 = (10-6) /
4 = 1.

Next, the slope K of the position signal POSN is calculated. This slope K
is calculated using the already determined average values T (N+) and T (N-) of the upper and lower time intervals of the reference voltage R, and the predetermined amplitude setting value V of the position signal shown in FIG. be done. That is, the amplitude setting value V is doubled, and V is the sum of the average values of the detection times of the upper and lower points of the reference voltage R.
It can be found by dividing by (N0) +T(N-))=(T1+T2).

As a result, the offset Vof (N) of the position signal POSN that is finally determined is the value obtained by multiplying the slope K of the position signal by the offset time interval To( (N).

In S4, the offset amount Vol for the position signal POSN is determined.
(N), the process advances to S5 and the position signal POS is calculated.
For Q, the calculation shown in the figure is performed in the same way to obtain the offset iiVol (Q).

Once the offset amounts Vof (N) and Vol (Q) are obtained for the position signals POSN and POSQ in this way, the process advances to S6, and from then on, the position signals POSN
and the offset fltVol (N), which has already been calculated from the sample data at the time of sampling POSQ,
Correction is performed to remove the offset by subtracting Vat (Q).

Incidentally, although the above embodiment described a two-phase position signal, 1
The same applies to phase position signals.

In addition, the calibration mode for measuring the offset amount of the position signal is set not only when the disk device is started up, but also periodically during idle time during operation, in order to calculate the offset amount over time. It may also be possible to ensure that changes can be removed.

[Effects of the Invention] As described above, according to the present invention, the amount of offset is detected from the position signal obtained by actually moving the head, and the offset amount of the position signal is automatically corrected after detection. It is possible to reliably eliminate track pitch errors and minimize track pitch errors, improve track servo control accuracy, and at the same time enable higher track density, which in turn greatly contributes to increasing storage capacity. Can be done.

[Brief explanation of the drawing]

FIG. 1 is a diagram explaining the principle of the present invention; FIG. 2 is a configuration diagram of an embodiment of the present invention; FIG. 3 is a functional block diagram of an MPU that performs offset removal of the present invention; FIG. 4 is a diagram showing the disk servo pattern and An explanatory diagram of successive signal waveforms; FIG. 5 is a timing chart of servo demodulation operation of the present invention; FIG. 6 is an explanatory diagram of two-phase position signals of the present invention. Fig. 7 is an operation flow diagram of the present invention; Fig. 8 is an explanatory diagram of two-phase position signals during offset detection of the present invention; Fig. 9 is an explanatory diagram of offset calculation of the present invention; Fig. 10 is a position signal of the conventional method. It is an offset explanatory diagram of. In the figure, 10: 12: 14 = 16: 18: 20: Disk medium (servo disk) Servo head Servo demodulation means (circuit) Head control means Cross time interval detection means Offset time interval detection means 22: Inclination detection means 24: Offset Calculation means 26.28: Register 30.32: Adder 34: AD converter 36: MPU 38: DA converter 40: Power amplifier 42: Head motor 44: AGC amplifier 46-1 to 46-4: Peak detector 48-1. 48
-2: Differential amplifier 50: Reference voltage generator

Claims (4)

[Claims] (1) A disk medium (10) recording a servo pattern in which the same amplitude pattern is repeated in the circumferential direction and different types of amplitude patterns are repeated in the radial direction in units of multiple tracks.
and; a servo head (12) that reads a servo pattern recorded on the disk medium (10) and outputs a servo signal; servo demodulation means (1) for creating a position signal indicating the radial head position with respect to
4) Head control means (16) that performs a seek operation to move the head in the radial direction based on the position signal from the servo demodulation means (14) and an on-track operation that causes the head to follow the track; In the servo control circuit of the disk device, when the servo head (12) is driven at a constant speed in the disk radial direction, the position signal obtained from the demodulation means (14) increases the reference level (R) in an upward direction. a first time interval (T1) from when the position signal crosses the reference level (R) until the next time it crosses it in the downward direction; and a second time interval (T1) from when the position signal crosses the reference level (R) in the downward direction until it next crosses it in the upward direction. cross time interval detection means (18) for detecting T2); offset time interval detection means for detecting an offset time interval (Tof) of the position signal based on the first and second time intervals (T1, T2); (20) and; slope detection means (22) for calculating the slope (K) of the position signal based on the first and second time intervals (T1, T2) and the prescribed amplitude value (V) of the position signal; and; said offset time interval (
an offset calculation means (24) for calculating and storing the offset amount (Vof) of the position signal by multiplying the slope (K) of the position signal by the slope (K) of the position signal; An offset removal method for a disk position signal, characterized in that the signal is corrected by the offset amount (Vof) and outputted to the head control means (16). (2) The cross time interval detection means (24) detects a plurality of first and second time intervals (T1i-T1n, T21-T
2. The offset removal method for a disk position signal according to claim 1, wherein each average value is calculated and output after detecting 2n). (3) The offset time interval detection means (20) calculates the offset time interval (Tof) by dividing the difference (T1-T2) between the first and second time intervals by 4. 1. The disk position signal offset removal method described in 1. (4) The slope means (22) doubles the amplitude voltage (V) at the first and second time intervals (T_1, T
2). The offset removal method for a disk position signal according to claim 1, wherein the slope (K) is calculated by dividing by the sum of 2).