JPH04337574A - Head position recognition method, head speed calculation method, and head movement speed control device - Google Patents

Abstract

PURPOSE:To enlarge the range of speed detection and the dynamic range of moving speed and to improve the performance of track seeking by providing four kinds of a servo pattern with the period of 24 tracks on servo sectors provided discretely in the direction of circumference on a recording medium. CONSTITUTION:To the sub-track code part of the servo sectors 2, four kinds of the servo pattern with the period 24 tracks are buried. This is read out by a head 5, the servo sectors existing discretely are detected by a servo information demodulator 8, by data found with an in-track position decoder 9, a positional information recognizing device 10, etc., head speed is found through a speed commander 11 and simultaneously, error between the output of a speed operation means 12 is calculated by an error amplifier to compensate by a compensator 14. Thus, the relative positional relation of the tracks where a data head passes through is detected, is corrected by one track or below, the range of the speed detection is enlarged, and also the dynamic range is enlarged and the track seeking is performed with higher speed and in shorter time.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a head position recognition method, a head speed calculation method, and a head movement speed control device when a data head accesses an arbitrary information track on a rotatable recording medium.

0002

2. Description of the Related Art Conventionally, in magnetic disk drives, a servo surface servo method has been widely adopted in which a head is positioned to a target data track based on servo information written on the entire surface of a recording medium. However, in this method, the data head is positioned based on a trusting relationship between the servo surface and the data surface, so off-track tends to occur due to various environmental changes, such as temperature changes within the device, which essentially increases the track density. The drawback was that it was difficult.

[0003] Recently, therefore, a data surface servo method has been attracting attention, which improves reliability during recording and reproduction by writing servo information on the data surface. One of them is the sector servo method. In this method, a servo sector for positioning is embedded in advance at the beginning of each sector on the data surface.
In this method, when an arbitrary data track is selected, the head is made to follow each data track based on servo information of the servo sector. However, with this method, the servo sector only has information for track following control, so in order to move the head at high speed, a separate position detector must be installed or head position information must be supplied from the servo surface. It has shortcomings. In addition, if you try to obtain sufficient position information even during high-speed access using only the servo information of the servo sector, the ratio of the servo sector to the data sector on the recording medium surface will increase, resulting in a significant decrease in storage capacity. are doing.

[0004] Therefore, a method has been proposed in which a track number is coded and recorded in a servo sector, and the head is accessed based on this information. (Japanese Unexamined Patent Publication No. 51-131607) This method calculates the average speed between sectors by discretely obtaining position information (address information) of the track that the head passes during seek, and compares it with the command speed. The speed is controlled depending on the situation. Therefore, it has been widely used recently because it is mechanically simple and has good cost performance even when the number of stacked recording media is small.

[0005] Also, truck position information (address information)
Accordingly, a method has been proposed to improve the speed control characteristics, which is a drawback of the method that performs speed control by determining the average speed between sectors (Japanese Patent Application Laid-Open No. 131607/1982). 39979).

FIG. 9 shows a specific example (servo pattern) of a servo sector proposed in the above-mentioned Japanese Patent Application No. 2-39979. It shows. 2 is a servo sector, and 3 is a data sector. This servo sector 2 includes a burst section 18 for obtaining an AGC signal,
Erase section 19 for detecting servo sectors, sub-track code section 20 for obtaining track address information
, position information 21 is provided to obtain information on the positional deviation of the data head from on-track during tracking control. The erase unit 19 is designed to have the maximum erasing time in the information track of the recording medium 1, and the position information 21 is composed of a burst signal α and a burst signal β. It is set by . Note that writing to servo sector 2 is prohibited.

Now, the sub-track code section 20 has a sync bit S indicating the start of the sub-track code section 20, a guard zone, a data zone, and three dibit patterns A, B, and C for determining the type of data zone.
, and a first zone discriminator 20a consisting of three-phase dibit patterns G, H, and I having a period of three tracks.
servo pattern 20b, two dibit patterns D with a period of 12 tracks and a deviation of 3 tracks.
, E, the second servo pattern 20c has a period of 6 tracks, and has at least 1 servo pattern between
．． A third servo pattern 20d consisting of a dibit pattern F having a five-track shift is embedded.

When performing a recording/reproducing operation on a data track, the data head 5 runs across two adjacent tracks with respect to the servo sector 2. Assuming that the data head 5 is now at the sixth track position on the recording medium, the waveform reproduced by the data head 5 will be as shown in FIG. That is, a predetermined reference signal is provided in the burst section 18, no signal is provided in the erase section, and
In the sub-track code section 20, the sync bit position S, the zone discrimination positions A, B, and C, the second servo pattern positions D and E, the third servo pattern position F, and the first servo pattern position G , H, and I, and furthermore, at the position of the position information 21, signals corresponding to the respective servo patterns are obtained at the burst signal α position and the burst signal β position. At this time, the data head 5 has two
Because it runs half astride two trucks,
If a pattern exists on only one side, the output will be approximately 1/2 that of a case where patterns exist on both sides.

In Japanese Patent Application No. 2-39979, using the reproduced waveforms A to I, D, E, F, G, H, and I, it is determined in which position of the servo pattern of 12 track periods the data head is located. Discloses a method and apparatus for recognizing whether a person is present or not. First, using the D, E, and F binary signals (for example, L, H, and H in the case of data head 5), the position where the data head is located is determined from track 12 to track 3 by means such as table drawing. Limited to trucks. Next, G, H, I,
After peak-holding the reproduced signal, G>H, H>I,
Binarized information with I>G (for example, H, L, H in the case of data head 5) is created. Then, using this binary signal, the above-mentioned limited 3
Further from the track, the position where the data head exists is limited to an area of 1/2 of the track width. Furthermore, the above G>
Based on the binary information of H, H>I, I>G, G, H, I
The signals showing the maximum and minimum peak hold values are selected (for example, in the case of the data head 5, the maximum is G and the minimum is H). Then, a signal indicating the minimum peak value (for example, H
), add an offset of approximately half the value when the peak hold value is at its maximum. Then, again, the signal whose peak hold value is the second value among the three peak hold values (for example, I) and the signal to which the above offset has been added (for example, H + offset), (H + offset) > I. Make a comparison. By this comparison, the above limited 1/2
Further from the track, the position where the data head exists is limited to an area of 1/4 of the track width. Result, D, E, F
, G, H, I signals, the position of the data head in the 12-track cycle servo pattern is divided into 48 (=12*4) sub-subs with a resolution of 1/4 of the track width. It is possible to recognize the track code.

The head positioning device disclosed in Japanese Patent Application No. 2-39979 uses the head position information recognition method described above to position the head with respect to a target track. In other words, when performing a head positioning operation with respect to a target track using speed control, it is possible to determine where the head is located in the servo pattern of 12 track periods from the reproduction signal from the servo pattern obtained each time it passes through a servo sector. Recognize the presence of animals. Then, each time it passes through a servo sector, the distance traveled by the head between samples is calculated from the head position detected one sample before and the current sampled head position, and the distance traveled by the head is divided by the sampling period. Based on this, the average moving speed of the head between samples is determined. When positioning the head to the target track by speed control, a target speed that allows the head to enter the target track stably is commanded for each servo sector in advance according to the relative distance to the target track. Therefore, the speed control system is
It is constructed by feeding back the difference between the target speed and the average moving speed of the head between the samples.

For example, during speed control, assume that the head position one sample before was located in the area of sub-track code 2 in a servo sector with a 12-track period, and is currently located in the area of sub-track code 20. . Then,
The average moving speed V between samples of the head at that time is given by (Formula 1) where the sub-track code of one sample before is STCo, the currently sampled sub-track code is STCn, the track pitch is Xtp, and the sampling period of the servo sector is TS. V = (STCn-STCo)*
(N*Xtp/48)/Ts. Now, Xtp=
12μm, Ts=300μsec, V=18c
The average speed of movement of the head in m/s is known. In this way, the speed control system is constructed by determining the average moving speed of the head, subtracting it from the target speed corresponding to the distance to the target track, and using this value as feedback information.

0012

[Problem to be solved by the invention] Japanese Patent Application No. 2-39979
The servo pattern consists of 12 track periods. Therefore, if the moving distance of the head between servo sectors exceeds 12 tracks, the average moving speed of the head between servo sectors cannot be calculated correctly.

For example, during speed control, when running at high speed, the head position one sample before is located in the area of sub-track code 2 in a servo sector with a 12-track period,
Assume that the current location is in the area of sub-track code 3. Then, the average moving speed V between samples of the head at that time is, since the moving distance between servo sectors is originally 48 sub-track codes + 1 = 49 sub-track codes.
V=49cm/s must be detected.

However, using equation 1, V=1c
m/s. That is, since an error calculation with respect to the target speed is performed based on incorrect speed detection, not only the speed control system cannot be configured correctly, but also a seek error may occur. Therefore, if the moving distance of the head between servo sectors is limited to 12 tracks, the maximum speed will be kept low, resulting in the problem of increasing the seek time to the target track.

For example, the track pitch Xtp=12
μm, and when the sampling period Ts is 300 μsec, the maximum speed Vmax is 48 cm/s. Normally, in the case of a 3.5-inch hard disk device, the average access time is more than 10 m.
The maximum speed is set to about 1 m/s due to the requirement of 1 m/s. Of course, it is possible to increase the maximum speed by shortening the sampling period Ts, but this increases the proportion of servo sectors on the recording medium and reduces the data recording capacity. Furthermore, if the track pitch Xtp is increased, the maximum speed can be similarly increased, but this reduces the number of tracks provided on the recording medium and reduces the data recording capacity. Therefore, limiting the movement distance of the head between servo sectors to 12 tracks poses a major problem in the detection range of the head movement speed during seek, and this poses a major problem in improving the performance of the entire head positioning device. This has become a problem.

[0016]

[Means for Solving the Problems] In order to solve the above-mentioned problems, the head position recognition method, head speed calculation method, and head movement speed control device of the present invention are provided with the following methods or configurations.

That is, the head position recognition method involves forming servo sectors discretely in the circumferential direction of a rotatable recording medium,
This servo sector includes a first servo pattern with a period of 3 tracks and at least 3 phases, a second servo pattern with a period of at least 24 tracks and a pattern of at least 2 phases with a deviation of at least 6 tracks from each other, and 12 A third servo pattern having a period of 6 tracks and consisting of a pattern of at least 2 phases with a shift of 3 tracks from each other, and a third servo pattern having a period of 6 tracks.
and a fourth servo pattern having a deviation of at least 1.5 tracks from the servo pattern, and the second servo pattern is used according to reproduction signals obtained from the four types of servo patterns on the recording medium. at most 1
Discriminate down to the 0 track range, use the third and fourth servo patterns to discriminate up to the 3 track range, and then
This method uses servo patterns to discriminate the head position down to an area of one track or less.

Further, in the head speed calculation method, discrete servo sectors are formed in the circumferential direction of each surface of a recording medium having a plurality of recording surfaces, and the servo sectors have a servo pattern of N track periods (N is an integer). The above servo pattern is divided into M sub-track codes (M≧N), and the sub-track code detected chronologically by the head every time it passes through a servo sector is divided into P sub-track codes (P is an integer) before the sub-track code. The head movement distance is calculated from the difference with the sub-track code detected in the servo sector passed through, and when calculating the head movement speed, the first offset is added to the head movement distance calculated in the servo sector passed P times before. A first head movement distance and a second head movement distance are created by subtracting a second offset from the head movement distance calculated at the servo sector passed P times before, and the second head movement distance is calculated at the current servo sector. When the head movement distance is larger than the first head movement distance, the value obtained by subtracting M from the calculated head movement distance is set as the true head movement distance,
In addition, when the head movement distance calculated in the current servo sector is smaller than the second head movement distance, the head movement is performed with the value obtained by adding M to the calculated head movement distance as the true head movement distance. This is a method of calculating speed.

Further, the head movement speed control device includes a rotatable recording medium, a servo sector according to claim 1 formed in the circumferential direction of the recording medium, and a data head capable of reproducing at least information on the recording medium. servo information demodulation means for extracting servo information included in the discrete servo sectors from the reproduced signal of the data head; 1/( of track width)
a head position information recognition means configured to be able to minutely recognize up to 2N); a speed command means for outputting a track access speed command according to the distance to the target track based on the output of the head position information recognition means; The track access control includes a speed calculation means for calculating the moving speed of the data head in the radial direction of the recording medium based on the output of the recognition means, and a positioner means for moving the data head to an arbitrary position in the radial direction of the recording medium. The apparatus has a configuration in which a signal based on a speed error between the command means and the speed calculation means is fed back to the positioner means.

[0020]

[Operation] The head position recognition method of the present invention embeds discrete servo sectors in the circumferential direction on a recording medium, and provides four types of servo patterns with a period of 24 tracks in the servo sectors. The structure is such that the average moving speed of the head between servo sectors can be correctly calculated up to a distance of 24 tracks between the servo sectors of the head.

For example, the track pitch Xtp=12
μm, and when the sampling period Ts is 300 μsec, the maximum speed Vmax can be set to 96 cm/s by setting the servo pattern to at least 24 track periods. Therefore, the limit for calculating the head movement distance between servo sectors is improved, and the speed detection range can be doubled compared to the conventional method. As a result, it is possible to achieve faster and shorter track seek (access) performance.

Further, the head speed calculation method of the present invention is a method of calculating the head movement speed based on the head movement distance from the servo sector passed P times before to the servo sector currently passing through. The first head movement distance is calculated by adding the first offset to the head movement distance calculated for the servo sector passed through in A second head movement distance is created, and the head movement distance calculated in the current servo sector is the first head movement distance.
and the second head movement distance, the head movement speed is calculated using a value obtained by subtracting M from or adding M to the calculated head movement distance as the true head movement distance. .

In this method, even if the servo sector is formed by a servo pattern with a short repetition period,
Estimate the range of the true head movement distance that should be found in the current servo sector based on the previously calculated head movement distance,
It has a function of correcting the head movement speed calculated by Equation 1 so that it becomes the true head movement speed. As a result, the head movement speed can be calculated without the upper limit of the head movement speed being limited by the repetition period of the servo pattern, making it possible to achieve faster and shorter track seek (access) performance. be.

Further, the head movement speed control device using the head position recognition method and head speed calculation method described above has a speed control device by providing four types of servo patterns having a period of at least 24 tracks in discretely formed servo sectors. It has the advantage of increasing the detection range and the ability to expand the dynamic range of head movement speed without being limited by the servo pattern repetition period. Therefore, faster and shorter track seek (access)
This has the effect of making it possible to realize performance.

[0025]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A head position recognition method, a head speed calculation method, and a head movement speed control device according to an embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows a specific example (servo pattern) of discrete servo sectors embedded in advance in information tracks on a rotatable recording medium 1 in an embodiment of the present invention. 2 is a servo sector, and 3 is a data sector. This servo sector 2 includes a burst section 18 for obtaining an AGC signal, and an erase section 19 for detecting the servo sector.
, a sub-track code section 20 for obtaining track address information, and a position information 21 for obtaining information on the positional deviation of the data head from on-track during tracking control. The erase unit 19 is designed to have the maximum erasing time in the information track of the recording medium 1, and the position information 21 is composed of a burst signal α and a burst signal β. It is set by . Note that writing to this servo sector is prohibited.

Now, the sub-track code section 20 has a sync bit S indicating the start of the sub-track code section 20, a guard zone, a data zone, and three dibit patterns A, B, and C for determining the type of data zone. ,
a first servo pattern 20b consisting of three-phase dibit patterns G, H, and I with a period of 3 tracks; and a first servo pattern 20b with a period of 24 tracks and 6
Two dibit patterns M and N with track deviation
A second servo pattern 20c consisting of a second servo pattern 20c, a third servo pattern 20d consisting of two dibit patterns D and E having a period of 12 tracks and a deviation of 3 tracks,
A fourth servo pattern 20d consisting of a dibit pattern F having a period of 6 tracks and having a deviation of at least 1.5 tracks from the third servo pattern is embedded.

Here, in FIG. 2, the second servo pattern 20c, the third servo pattern 20d, and the fourth servo pattern 20d, each consisting of two dibit patterns, are arranged on one side of the first servo pattern 20b. There are, but
The first, second, third, and fourth servo patterns may be formed in any order. When performing a recording/reproducing operation on a data track, the data head 5 runs across two adjacent tracks with respect to the servo sector 2.

If the data head 5 is now located at the sixth track position on the recording medium, the waveform reproduced by the data head 5 will be as shown in FIG. That is, in the burst section 18, a predetermined reference signal, in the erase section, no signal, and in the sub-track code section 20, at the sync bit position S, at the zone discrimination positions A, B, and C, at the second servo pattern position. M, N,
At the third servo pattern positions D and E, at the fourth servo pattern position F, at the first servo pattern positions G, H, and I, and at the position of position information 21, the burst signal α position and the burst signal position β At each position, a signal corresponding to each servo pattern is obtained. At this time, the data head 5 travels across two tracks in half in the servo sector 2, so when a pattern is present on only one track, the output is approximately 1/2 as wide as when patterns are present on both sides. It becomes 2. In Figure 3, positions S, A, B,
A reproduced signal is obtained with an output of 1 at C, M, N, E, F, G, and I, an output of 0 at positions D and H, and an output of 1/2 at burst positions α and β. It goes without saying that this reproduced signal varies depending on the track position of the data head 5 on the recording medium.

In the present invention, the first
to fourth servo patterns 20b, 20c, 20d
, 20e, the track position of the data head 5 with respect to the recording medium is detected with high precision as follows, and the head positioning is performed by performing a seek operation to the selected target track.

FIG. 3 shows the ideal output state transition of the reproduced waveform when the data head 5 moves the servo sector shown in FIG. 2 at low speed from the inner circumference to the outer circumference of the recording medium. Since the data head 5 has a finite width, the slope of each reproduced signal in FIG. 3 is determined by the output value according to the ratio of the dibit pattern to the head width when the head crosses each dibit pattern. . Further, the numbers from 00 to 23 in FIG. 4 correspond to the track numbers with a 24-track period shown in FIG. It shows.

Here, the reproduced signals M, N, D, E, and F obtained from the second, third, and fourth servo patterns are binarized at a preset level, and then converted to the first servo pattern. The reproduced signals G, H, and I obtained from the pattern peak-hold each reproduced output, and then compare the respective values to obtain binary information, and reproduce the reproduced signals M, N, D, E, F, G, H. , the above two binarized information from I are the first binarized information,
Next, according to the value of the first binarized information, the G
, H, and I signals, a preset offset is added to the signal holding the lowest value, and the comparison with the corresponding signal is performed again to obtain the second binary information,
The position of the data head 5 on the servo track is determined by the second binary information and the second binary information.
/(22) is recognized minutely.

First, the second servo patterns M and N each have a period of 24 tracks, have a deviation of 6 tracks, and 6 tracks overlap each other, and the M signal and N signal are obtained from the reproduced signal. . Therefore, if the reproduction signal is ideal, it is possible to limit 6 tracks out of 24 tracks only by reproduction output of the M signal and the N signal. However, because the head has a finite width and the playback output does not change digitally, and the playback waveform is affected by the recording medium and the response characteristics of the data head, the limitable track area is six tracks. Spread more. In other words, as shown in FIG. 3, the area that can be limited only by the binarized M signal and N signal is 7 tracks, including the transition from H to L and from L to H signal of each reproduced signal. becomes.

Of course, the repetition period of the M signal and the N signal is set to 2.
If the number of tracks is larger than four, the area of tracks that can be defined by the M signal and the N signal will be expanded. For example, if the repetition period of the M signal and the N signal is set to 36 tracks, and the structure is configured such that there is a shift of 9 tracks and the 9 tracks overlap each other, the area of tracks that can be limited only by the M signal and the N signal is 1.
It becomes 0 track. Incidentally, since the repetition period of the next D signal and E signal is 12 tracks, the repetition period of the M signal and the N signal may be set to a multiple of 12, for example, 48 or 60. However, when the repetition period of the M signal and the N signal is set to 48 tracks, the M signal and the N signal
The area that can be limited by the signal is from H to L of each reproduced signal.
Including the transition from L to H signal and the transition from L to H signal, there will be 13 tracks. Therefore, it is no longer possible to further finely recognize the head position using the next D, E, F signals. As a result, the repetition period of M signal and N signal is 3
Maximum of 6 tracks.

After limiting from 24 tracks to 7 tracks using the binarized M signal and N signal with a 24-track period (in the case of a 36-track period, it is possible to limit from 36 tracks to 10 tracks), Furthermore, D, E, F
, signals are used to limit the position of the head relative to the servo pattern.

The third servo patterns D and E each have 12
The period is the track, and there is a shift of 3 tracks, and 3
The tracks overlap each other, and a D signal and an E signal are obtained from their playback outputs. Therefore, if the reproduced signal is ideal, it is possible to limit three tracks out of the 12 tracks only by the reproduced output of the D signal and the E signal. However, the head has a finite width, so the reproduction output does not change digitally, and the reproduction waveform is affected by the recording medium and the response characteristics of the data head. It is difficult to limit. Therefore, with a period of 6 tracks, D pattern or E
A fourth servo pattern F was provided which overlapped with the pattern by 1.5 tracks. Therefore, the fourth servo pattern F always becomes the "L" level or the "H" level at the slope portion of the D signal or the E signal.

As described above, in a servo sector with a period of 24 tracks, the M signal, N signal, and D signal are transmitted for each servo track.
The binarized information of the signal, E signal, and F signal is shown in (Table 1).

[0038]

[Table 1]

The "*" mark indicates whether the binary output is at the "L" level or "H" depending on the position of the data head at the time of signal detection.
This is a region in which it is unclear whether the level will be ``L'' or ``H''.As can be understood from (Table 1), when the level of either signal D or signal E is uncertain, the level of signal F is determined to be ``L'' or ``H''. Therefore, if the levels of signal M, signal N, signal D, signal E, and signal F are detected, it is possible to determine in which three of the 24 tracks the data head 5 is located. Can be limited.

[0040] In Table 1, since the M signal and the N signal have a 24-track period, the 5-track period of M, N, D, E, F,
All cases expressed in bits (e.g. HLLLL, etc.)
does not exist in (Table 1).

Next, it is determined from the reproduced signals G, H, and I obtained from the first servo pattern at which position within the limited three tracks the data head 5 is located. Similar to FIG. 3, FIG. 4 shows the ideal reproduction output of signals G, H, and I when the data head 5 moves the servo sector shown in FIG. 1 at low speed from the inner circumference to the outer circumference of the recording medium. It shows the output state transition. In Figure 3,
The figure shows an enlarged view of the data head 5 moving from servo track c to g, and signals G, H, and I.
are shown overlapping each other. Also, the peak values of the outputs of the signals G, H, and I are shown in a normalized state (the maximum is set to 1). Now, the detection code of signals M, N, D, E, F is “H”
Assuming that it is “L”, “H”, “H”, and “L”, (Table 1)
As can be understood from the above, the data head is located somewhere among tracks d, e, and f within the servo sector. Assuming that the peak values of the reproduced signals from signals G, H, and I are 22, 23, and 24, respectively, as shown in FIG.
G signal value > H signal value, H signal value > I signal value, I
It can be seen that the value of the signal<the value of the G signal. That is, the value of the G signal>the value of the H signal, the value of the H signal>the value of the I signal,
When the value of the I signal is compared with the value of the G signal, the answer is "H", "H", and "L". From this answer, it is determined that the peak values of G, H, and I are larger in the order of G, H, and I. As a result, as can be seen from FIG. 4, the position of the data head is within the e track of the three tracks d, e, and f, since the G signal is the largest, and the value of the H signal > the I signal. From the value of , it is determined that the area is located in the area e2 (shown in FIG. 4).

Next, the first binary information (M signal, N signal, D signal, E signal, F signal, G signal, H signal, I signal are "H", "L", "H", respectively) , “H”, “L”, “H”
”, “H”, “L”), an offset is added to the I signal with the lowest peak value among the G, H, and I signals.
The second offset value is equivalent to about 0.5 when the maximum peak values of G, H, and I are normalized to 1. 0 to I signal
．． After adding an offset of 5, the peak value of the I signal moves from position 24 to 25 in FIG. After that, when the comparison is made again such that the value of the H signal is greater than the value of the I signal, the answer is "L". Of this answer,
H signal value > I signal value (after adding 0.5 offset)
A comparison of the data head is useful for determining the more detailed position of the data head. In other words, the value of the H signal > the value of the I signal (
After adding an offset of 0.5), if the answer is "H", the peak value of the H signal is 0.5 or more larger than the peak value of the I signal, and the answer is "L". Then, the difference between the peak value of the H signal and the peak value of the I signal is 0.
．． It will be smaller than 5. That is, it is possible to determine whether the position of the data head is in the first half (e03) or the second half (e04) of the area e2. This time, the answer for H signal value > I signal is "L", so the difference between the peak value of H signal and the peak value of I signal is less than 0.5, so as can be seen from Figure 4, the data It is determined that the head position is located in area e03 of area e2 shown in FIG. In this way, the second binary information allows the relative positional relationship of the data head to the track in the servo sector to be determined in detail down to 1/(22) of the track width.

Furthermore, by adding an offset of -0.25 to the I signal, the value of the H signal>the value of the I signal (0.5-0.2
5), and the third 2
By creating the value information, the relative positional relationship of the data head to the track in the servo sector can be changed to 1 of the track width.
/(23) can be determined in detail.

As described above, by adding the offset given by ±1/2N (formula 2) to the G signal, H signal, or I signal according to the Nth binarized information, the track of the data head is adjusted. It becomes possible to determine the relative positional relationship to 1/(2N) of the track width in detail.

As described above, G signal value > H signal value, H signal value > I signal value, I signal value > G signal value according to the G signal, H signal, and I signal of 3 track periods in the servo sector. The first binarized information of , is summarized in (Table 2). Here, the alphabets from a to l, a1, a2 to l1, and l2 are the same as those shown in FIG. In addition, if the binarized information is "H", "H", "H", or "L", "L", "L"
”, so it does not exist in the table.

[0046]

[Table 2]

For example, from Table 1, it is assumed that the data head is located somewhere among tracks d, e, and f within the servo sector. Then, if the data head is located at the position shown in Figure 4, the answers to G>H, H>I, and I>G are "
Since the values are H""H" and "L" (Table 2), it is determined that the data head is located in the area e2.

Next, the second binarized information is created in order to discriminate in detail down to 1/4 of the track width. Table 3 shows which signals are to be offset.

[0049]

[Table 3]

G>H, H in the first binary information signal
If >I, I>G is "H", "H", and "L", the peak values of the signals G, H, and I will be larger in the order of G, H, and I, and the offset (0.5) It is the I signal that is added. Using (Table 3), G>H, H>I, I>
If the answer to G is "*", "H", or "L", the name of the signal to which the first offset is added is the I signal, which is consistent with the above result. Therefore, by using (Table 3), the signal name to which the first offset is added can be determined. Furthermore, after adding an offset, G>H, H>I,
By comparing I>G, second binarized information is obtained, making it possible to determine the position of the data head up to 1/(22) of the track width. (Table 4) shows the results of region discrimination based on the second binarized information. Note that the "$" mark indicates the signal name with an offset added, and the "*" mark indicates no relation. In the case of the above example, the second meaningful binary information, that is, H>I, becomes "H", so
From (Table 4), it is determined that the data head is located in the e03 area.

[0051]

[Table 4]

Furthermore, in order to determine the position of the data head up to 1/(23) of the track width, the first offset is determined depending on whether the answer of the second binary information is "H" or "L". A second offset of 0.
You can either add 25 or subtract 0.25. Furthermore, 0.25 may be added to the other compared value to create the second binarized information. For example, in the case of the above example, from the peak value of the I signal to which 0.5 is added,
．． After subtracting 25, the comparison of H>I is performed again, or 0.25 is added to the H signal and the comparison of H>I is performed again. If the answer is "H", it is determined that the data head is located in the second half of the e03 area, and if the answer is "L", it is determined that the data head is located in the first half of the e03 area, and the data head position is extended to 1/(23) of the track width. It becomes possible to determine the In any case, an offset of ±(1/(2N)) of the maximum value of the reproduced signal is added to the lowest peak value of the reproduced signals G, H, and I, so that G>H, H> By proceeding with the comparison of I and I>G, it becomes possible to determine the position of the data head up to 1/(2N) of the track width.

By using the M signal and the N signal as described above, the dynamic range of position recognition can be expanded to 24 track periods. Needless to say, if the M signal and N signal are formed at a period of 36 tracks, the dynamic range of fine position recognition can be expanded to 36 tracks.

Therefore, for example, if the track pitch is
When p=12 μm and the sampling period Ts=300 μsec, the maximum speed Vmax can be set to 96 cm/s by setting the servo pattern to at least 24 track periods. Also, the servo pattern must be at least 36.
By setting the track period, the maximum speed Vmax = 14
It can be set to 4cm/s. Therefore, the limit for calculating the head movement distance between servo sectors is improved, the speed detection range can be doubled or tripled compared to the conventional method, and high-speed and short-time track seek (access) performance can be realized.

Furthermore, compared to the advantage of realizing high-speed and short-time track seek performance by increasing the maximum speed, the servo patterns M and N proposed in Japanese Patent Application No. 2-39979 are The disadvantage of increased servo area caused by the addition is negligible. This is because the M signal and N signal are included in the servo pattern.
This is because the reduction in data area caused by adding a signal is at most about 0.1 to 0.2% in terms of capacity, which can be absorbed in the design.

Next, the speed calculation method of the present invention will be explained. FIG. 5 is a flowchart illustrating the speed calculation method of the present invention. In Figure 5, the head movement speed is V, the track pitch is Xtp, and the servo sector sampling period is T.
S, the period of the servo pattern is N, the increment of the servo pattern with the above N track period is M, the sub-track code of the P previous servo sectors is STCo, from the (2*P) previous servo sectors to the P previous servo sectors. The head movement distance is XSTCo, the currently sampled sub-track code is STCn, and the head movement distance from the P previous servo sectors to the current servo sector calculated using Equation 1 is XSTCn.
, the true head movement distance from the P previous servo sector to the current servo sector is RXSTCn, the first offset is A*P, and the second offset is B*P.

First, in step S100, each time the head passes through a servo sector, a sub-track code STCn is detected, which allows recognition of the position of the head in the servo pattern of N track periods. Then, step S1
At step 01, the subtrack code STCn detected in the current servo sector is subtracted from the subtrack code STCo of the P servo sectors before to find the head movement distance XSTCn between the P servo sectors.

Next, it is checked whether the head movement distance XSTCn≧0 determined in step S102. This is P
This is to prevent the head movement distance from becoming a negative value when the sub-track code detected in the previous servo sector is a large value and the currently detected sub-track code is a small value. If XSTCn<0, in order to make the head movement distance obtained by the above simple subtraction a positive value, perform the operation in step S103, that is, XSTCn=X
Perform STCn+M operation.

Furthermore, it is determined in step S104 whether the head movement distance determined in step S101 or step S103 is within the upper limit of the true head movement distance. In other words, in this operation, the true head movement distance from the P previous servo sector to the currently passing servo sector is the head movement distance XSTCo from the (2*P) previous servo sector to the P previous servo sector. This is based on the idea that the value is within the sum of the acceleration of the servo actuator and the change in head movement distance (A*P) determined by the cycle between servo sectors. If XSTC
If o+(A*P)≧XSTCn is satisfied, proceed to the next step; otherwise, after subtracting M from XSTCn, the operation of step S104 is repeated again.

In step S106, it is determined whether the head movement distance obtained in step S101 or step S103 is greater than or equal to the lower limit of the true head movement distance. That is, in this operation, as in step S104, the true head movement distance from the P previous servo sectors to the currently passing servo sector is the head movement distance from the (2*P) previous servo sectors to the P previous servo sectors. This is based on the idea that the movement distance XSTCo is greater than or equal to the sum of the head movement distance change (B*P) obtained from the acceleration of the head movement actuator and the cycle between servo sectors. For example, if XSTCn≧XSTCo+(B
If *P) is satisfied, proceed to the next step; otherwise, after adding M to XSTCn, the operation of step S106 is repeated again.

As a result, since the head movement speed (XSTCn) that has passed through step S106 can be determined to be the true head movement speed, the true head movement speed RXSTCn and XSTCn are replaced in step S108. Then, in step S109, the head movement speed is calculated, and in step S110, the head movement distance (XSTCo) from the (2*P) previous servo sector to the P previous servo sector is calculated.
and the current true head movement distance (RXSTCn) are exchanged in order to calculate the head movement speed again in the next servo sector.

The head speed calculation method of the present invention has been described above, and will be explained in more detail using a specific example. For example, assume that a servo pattern of 12 track periods (N=12) is divided into 48 sub-track codes from 0 to 47 (M=48). Also, track pitch Xtp=12
μm, sampling period Ts=300 μsec, and P=1. Then, the period of the servo pattern becomes 144 μm, and the length of one sub-track code becomes 3 μm. If the acceleration of the head movement actuator is 28G at maximum, the change in the head movement distance that can be moved between servo sectors is approximately 12.3 μm, which is equivalent to approximately 4 subtrack codes. Therefore, the first offset A and the second offset B are set to about 6 sub-track codes in consideration of margin.

[0063] Now, suppose that the head is moving at a speed of about 45 cm/s, the head position one sample ago was in the area of sub-track code 2 in a servo sector with a 12-track period, and the current position is in the sub-track code 2 area. Assume that it is located in the area of track code 3. Then, the track pitch Xt
In the case where p=12 μm and the sampling period Ts=300 μsec, if Equation 1 is used, V=1 cm/s. However, normally, the speed should change smoothly, and it is unlikely that the speed will suddenly change to 1 cm/s with a short servo sector period of about 300 μsec and an acceleration of about several tens of Gs. Therefore, the speed detection method shown in FIG. 5 is used. In step S101, XSTCn=1 subtrack code is calculated, and in step S104, XSTCo
+A(45+6)≧XSTCn(1), which satisfies the condition, so the process advances to step S106. Step S1
In 06, XSTCn(1)≧XSTCo-B(45-
6) is not satisfied, step S107XSTC
Perform the operation n (49) = XSTCn (1) + M (48) to determine the head movement distance XSTCn with the sub track code.
=49 is obtained. As a result, a true head movement speed of 49 cm/s can be obtained in step S108.

Furthermore, even if the head is moving at a speed of about 95 cm/s, step S106
In the determination, first perform the operation in step S107, and
Although Cn=49, the determination is made again in step S106, but if XSTCn(49)≧XSTCo-B(95-
6), and since the conditions are not met, the operation in step S107 is performed again. Result, step S107
XSTCn(97)=XSTCn(49)+M(48
) is performed to obtain XSTCn=97. As a result, a true head movement speed of 97 cm/s can be obtained in step S109.

[0065] Also for negative speeds, step S10
This method can be handled by providing the determination in Step 4 and the operation in Step S105.

As described above, this method estimates the range of the true head movement distance that should be calculated for the current servo sector based on the previously calculated head movement distance, even if the servo sector is formed by a servo pattern with a short repetition period. However, it has a function of correcting the calculated head movement speed so that it becomes the true head movement speed. As a result, the head movement speed can be calculated without the upper limit of the head movement speed being limited by the repetition period of the servo pattern, making it possible to achieve faster and shorter track seek (access) performance. be.

It goes without saying that the head speed calculation method described above is effective even if the sub-track codes detected in time series are from servo sectors of different recording and reproducing surfaces that are rotating in the same way.

Next, a description will be given of a head movement speed control device using the four types of servo patterns each having a period of 24 tracks as described above.

FIG. 6 is a basic block diagram of a head movement speed control device in the first embodiment of the present invention. 1 is a rotatable recording medium such as a magnetic disk, which is rotated by a spindle motor (not shown). 2 is a servo sector discrete in the circumferential direction in which servo information of 24 track periods is recorded, which is embedded (recorded) in advance on the surface of the recording medium;
3 is a data sector in which data is recorded. 4 is an information track provided on the surface of the recording medium. 5 is a data head that can read and write information on this information track, 6 is a VCM positioner (voice coil motor) that moves (accesses) this data head to a selected information track, and 7 is amplified the reproduced signal from the data head. 8 is a servo information demodulator that detects servo sectors that exist discretely from the reproduced signal and extracts the information; 9 is a servo information demodulator that detects servo sectors that exist discretely from the reproduced signal and extracts the information;
The in-track position decoder 10 detects a dynamic range of two tracks, and uses the M signal to the I signal with a 24-track period or 36-track period to determine the relative positional relationship of the data head to the information track by 1/(of the track width). A head position information recognition device 11 recognizes and discriminates up to 2N), and 11 determines a target speed according to the distance to the target track or the number of tracks to the target track during track access control to move (access) the data head to the selected information track. It is a speed command device that commands. The speed command device is usually realized by a ROM table or the like, but it may be one that calculates a function and outputs it every time the distance to the target track or the number of tracks to the target track is measured. 12 is a speed at which the moving speed of the data head in the radial direction of the recording medium is calculated based on the head speed calculation method described above by inputting high-resolution head position information from the head position information recognition device every time the data head passes a servo sector. The calculation means 13 is an error amplifier that calculates an error between the outputs of the speed command device 11 and the speed calculation means 12. The output of the error amplifier 13 is supplied via a compensator 14 and a switch 16 to a current driver 17 that supplies current to the VCM positioner according to the output of the compensator 14, thereby forming a track access control loop. A tracking control loop for causing the data head to follow the selected information track is constructed by supplying the output of the intra-track position decoder to the current driver 17 via the compensator 15 and switch 16.

As can be understood from the basic block diagram of FIG. 6, the head movement speed control device equipped with the head position recognition method and head speed calculation method of the present invention is capable of recognizing the position and speed during track access (seek) of the data head. It is concerned with the control method, and is mainly concerned with the control method of 24 tracks or 36 tracks discretely embedded in the circumferential direction on the recording medium.
It is involved in a head movement speed control device that detects a servo pattern of a track period and the relative positional relationship of a track through which a data head passes from that pattern to one track or less, and is realized by the servo pattern and speed calculation means. . As a result, four types of servo patterns with a period of at least 24 tracks are provided in discretely formed servo sectors, which has the advantage of increasing the speed detection range, and the head movement speed is not limited by the servo pattern repetition period. It also has the advantage of being able to expand the dynamic range. Therefore, it becomes possible to realize faster and shorter track seek (access) performance.

FIG. 7 is a block diagram illustrating in more detail the head position information recognition device 10 and servo information demodulator 8 of the head movement speed control device in one embodiment of the present invention. The signal detected from the recording medium by the data head 5 is amplified by the preamplifier 7 and then sent to the servo sector 2.
The output value is transmitted to the AGC amplifier which normalizes the output value using the burst section 18 etc. embedded in the output value. The servo information demodulator 18 includes an AGC amplifier 38, a binarization circuit 39 that binarizes the signal standardized by the AGC amplifier 38 at a predetermined threshold, and a An erase part detector 40 detects the erase part 19 with a long interval, starts a counter at the same time as detecting the erase part 19, and starts a counter to detect the signals A to I of the sub-track code 20 embedded in the servo sector 2 and the inside of the track. It is composed of a sector counter and a gate generator 41 that generate gates for detecting α bursts and β bursts indicating position information, and also generate gates for discriminating data sector 3 and the next servo sector. The in-track position decoder 9 receives gate commands from the sector counter and gate generator 41, receives only α bursts and β bursts indicating in-track position information from the AGC amplifier, and performs the calculation (α burst - β burst). , detects the in-track position of the data head during tracking control. The head position information recognition device 10 also receives gate commands from the sector counter and gate generator 41, and receives signals A, B, C, and D among the signals binarized by the binarization circuit 39.
, E, F out of the signals standardized by the AGC amplifier 38 after receiving gate commands from the storage circuit 36 and the sector counter and gate generator 41. , a peak holder G26, a peak holder H27, and a peak holder I28 that hold the peak values of I, respectively, and a comparator 29 that compares the peak value of the peak holder G26 and the peak value of the peak holder H27.
, a comparator 30 that compares the peak value of the peak holder H27 and the peak value of the peak holder I28, a comparator 31 that compares the peak value of the peak holder I28 and the peak value of the peak holder G26, and comparators 29, 30, A first latch 32 holds the binarized information of 31 and constitutes the first binarized information together with the binarized information of the storage circuit 36; H,I
Which of the following should be added the first offset (Table 3)
Offset decoder 33 determined according to the decoding method of
, an offset adder 34 that adds a predetermined offset value to any one of the peak holders G, H, and I according to a command from the offset decoder 33; and a peak holder corresponding to the offset added by the offset adder 34. A second latch 35 holds the comparison result with the peak holder and forms second binary information, and further adds a second offset according to the result of the second binary information. Instruct the offset adder to select the peak holder to be added and the second offset value to be added,
The peak holder to which the second offset value has been added is compared with the corresponding peak holder, the third binary information is formed in the third latch, and the above operation is repeated to obtain the N-th peak holder. Forming binary information in the Nth latch 38,
As a result, the head position information determination element 37 uses the binarized information from the first latch to the Nth latch to determine the relative positional relationship of the data head to the servo track in detail down to 1/(2N) of the track width. configured.

With the configuration described above, the head position information recognition device 10 determines the relative positional relationship of the data head to the servo track by 1/(2N) of the track width.
It becomes possible to make detailed determinations. Therefore, at the time of track access control, the moving speed of the data head can be recognized with a resolution N times higher than that of the conventional technology, making highly accurate speed control possible. The head position information determination element 37 is as shown in (Table 1)
), (Table 2), (Table 3), and (Table 4) may be performed by hardware using a ROM table or the like, or by software using a μCPU or the like. It's okay.

[0073] At this time, the head position information discriminating element 37
can be constructed using only hardware such as a ROM table, but may also be constructed using hardware such as a 1-chip μCPU and software.

[0074]

As described above, the head position recognition method of the present invention embeds discrete servo sectors in the circumferential direction on a recording medium, and the servo sectors have four types of servo patterns having a period of at least 24 tracks. is provided, and the moving distance between the servo sectors of the head is 2 during speed control.
The system is configured to correctly calculate the average moving speed between servo sectors of the head up to four tracks.

For example, if the track pitch Xtp=12
μm, and when the sampling period Ts is 300 μsec, the maximum speed Vmax can be set to 96 cm/s by setting the servo pattern to at least 24 track periods. Furthermore, by setting the servo pattern to a period of 36 tracks, the maximum speed Vmax can be set to 144 cm/s. Therefore, the head movement distance calculation limit between servo sectors has been improved, and the speed detection range has been doubled or tripled compared to the conventional method.
It is possible to double it. As a result, it is possible to achieve faster and shorter track seek (access) performance. In addition, as mentioned in the explanation of the embodiment, newly added servo patterns M and N
The increase in the servo area caused by this is at most about 0.1 to 0.2% in terms of capacity, which is within the range that can be absorbed by design.

The head speed calculation method of the present invention is a method of calculating the head movement speed based on the head movement distance from the servo sector passed P times before to the servo sector currently passed through. The first offset is calculated by adding the first offset to the head movement distance calculated in the servo sector.
A second head movement distance is created by subtracting the second offset from the head movement distance calculated at the servo sector passed P times before, and the head movement calculated at the current servo sector is created. When the distance is not between the first and second head movement distances, the head movement speed is calculated using the value obtained by adding M to the calculated head movement distance as the true head movement distance. .

[0077] In this method, even if the servo sector is formed by a servo pattern with a short repetition period,
Estimate the range of the true head movement distance that should be found in the current servo sector based on the previously calculated head movement distance,
It has a function of correcting the head movement speed calculated by Equation 1 so that it becomes the true head movement speed. As a result, the head movement speed can be calculated without the upper limit of the head movement speed being restricted by the repetition period of the servo pattern, resulting in faster and shorter track seek (access) performance. be.

Further, the head movement speed control device using the head position recognition method and head speed calculation method described above has a speed control device by providing four types of servo patterns having a period of at least 24 tracks in discretely formed servo sectors. It has the advantage of increasing the detection range and the ability to expand the dynamic range of head movement speed without being limited by the servo pattern repetition period. Therefore, faster and shorter track seek (access)
This has the effect of making it possible to realize performance.

[Brief explanation of drawings]

[Fig. 1] Specific servo pattern diagram of a servo sector in the first embodiment of the present invention.

[Figure 2] Reproduction waveform diagram when the data head crosses the servo sector in Figure 1

[Fig. 3] Reproduction output state diagram when the data head moves at low speed over the servo sector in the same embodiment

[Fig. 4] Reproduction output state diagram of signal G, signal H, and signal I when the data head moves at low speed over the servo sector in the same embodiment.

FIG. 5 is a flowchart explaining the head speed calculation method of the present invention.

FIG. 6 is a basic block diagram of a head positioning device in the first embodiment of the present invention.

[Fig. 7] Block diagram of the servo information demodulator and head position information recognition means of the same embodiment.

[Figure 8] Specific servo pattern diagram of conventional servo sector

[Figure 9] Reproduction waveform diagram when the data head crosses the servo sector in Figure 8

[Explanation of symbols]

1 Rotatable recording medium 2 Servo sector 3 Data sector 4 Information track 5 Data head 6 VCM positioner 8 Servo information demodulator 9 In-track position decoder 10 Head position information recognition device 11 Speed command device 12 Speed calculation means 17 Current driver 18 Burst section 19 Erase section 20 Sub-track code 21 Position information 22 Peak hold value of reproduced signal G 23 Peak hold value of reproduced signal H 24 Peak hold value of reproduced signal I 25 Peak hold value after adding offset 0.5 to reproduced signal I 26, 27 , 28 Peak holder 29, 30, 31 Comparator 32 First latch 33 Offset decoder 34 Offset adder 35 Second latch 37 Head position information determination element 39 Binarization circuit 40 Erase section detector

Claims (6)

[Claims] 1. Servo sectors are formed discretely in the circumferential direction of a rotatable recording medium, and each servo sector has a first servo pattern of at least three phases having a period of 3 tracks, and a first servo pattern of at least 3 phases having a period of at least 24 tracks. a second servo pattern having a period of 12 tracks and a pattern of at least 2 phases and having a deviation of at least 6 tracks;
A third servo pattern consisting of a phase pattern, and a fourth servo pattern having a period of 6 tracks and having a deviation of at least 1.5 tracks from the third servo pattern, According to reproduction signals obtained from four types of servo patterns, a second servo pattern is used to discriminate up to a range of at most 10 tracks,
A head position recognition method in which the third and fourth servo patterns are used to discriminate up to a range of three tracks, and the first servo pattern is used to discriminate the head position to a range of one track or less. 2. The head position recognition method according to claim 1, wherein each of the first, second, third and fourth servo patterns is formed including at least one dibit pattern. 3. Discrete servo sectors are formed in the circumferential direction of each surface of a recording medium having a plurality of recording surfaces, a servo pattern having an N track period (N is an integer) is provided in the servo sector, and the servo sector is provided with a servo pattern of N track periods (N is an integer). The pattern is divided into M subtrack codes (M≧N), and each time the head passes a servo sector, the subtrack code is detected in chronological order by the head, and the servo sector passed P times before (P is an integer) is detected. The head movement distance is calculated from the difference with the sub-track code, and when calculating the head movement speed, the first head movement is calculated by adding the first offset to the head movement distance calculated in the servo sector passed P times before. A second head movement distance is created by subtracting the second offset from the head movement distance calculated at the servo sector passed P times before, and the head movement distance calculated at the current servo sector is calculated. , when it is larger than the first head movement distance, the value obtained by subtracting M from the calculated head movement distance is set as the true head movement distance, and the head movement distance calculated in the current servo sector is When the head movement distance is smaller than the second head movement distance, the head movement speed is calculated using a value obtained by adding M to the calculated head movement distance as the true head movement distance. 4. The first offset and the second offset are created based on the moving distance of the head that can be moved between the P servo sectors when accelerating and decelerating with the acceleration generated by the head moving actuator. Head speed calculation method described. 5. A rotatable recording medium, a servo sector according to claim 1 formed in the circumferential direction of the recording medium, a data head capable of reproducing at least information on the recording medium, and a reproduction signal of the data head. servo information demodulation means for extracting servo information included in the discrete servo sectors from the data head; and a head position configured to recognize the relative positional relationship between the data head and the track based on the head position recognition method according to claim 1. information recognition means; speed command means for outputting a track access speed command according to the distance to the target track based on the output of the head position information recognition means; and a positioner means for moving the data head to an arbitrary position in the radial direction of the recording medium, and track access control is based on a speed error between the speed command means and the speed calculation means. A head movement speed control device configured by returning a signal to the positioner means. 6. The head position information recognition means configured to be able to minutely recognize the relative positional relationship between the data head and the track down to 1/(2N) of the information track width detects the peak value of the reproduced signal amplitude of each servo pattern. at least two peak holder elements that store each, at least one comparison element that compares the at least two peak values, and a first latch that stores the output of the comparison element and holds the first binarized information. an offset decoder element that determines which peak holder element to apply an offset to according to the contents of the first latch element, and an offset that adds an offset to a predetermined peak holder element based on a command of the offset decoder element. The additional element, the peak holder element to which the offset is added, and the peak holder element to which the offset is not added are compared again using the comparison element, and the result is temporarily stored to hold the second binarized information. 2 latch elements, N latch elements that latch the binarized information up to the Nth latch element, and
Using the contents up to the latch element, the relative positional relationship of the data head to the track is determined by 1/(2N) of the information track width.
6. The head movement speed control device according to claim 5, further comprising a head position information discriminating element that detects up to ).