JPH0322672B2 - - Google Patents

Description

[Detailed Description of the Invention] [Technical Field of the Invention] The present invention is directed to positioning a head efficiently and with high precision in accordance with a servo signal on a recording medium in which a servo signal is embedded in a predetermined area of a data surface. The present invention relates to a head positioning method that can be used.

[Technical background of the invention]

In recent years, so-called floppy disk devices using flexible magnetic disk recording media (hereinafter referred to as "floppy disks") that are inexpensive and easy to handle have become widespread. In such a floppy disk device, head positioning is generally performed by an open loop method using a stepping motor.

However, with the recent progress in high-density recording media technology, the conventional head positioning method suffers from chuck errors when replacing disks, anisotropic expansion and contraction of the disk itself due to temperature and humidity, mechanical errors in the stepping motor, etc. A new problem has arisen in that it is impossible to avoid off-track caused by. For this reason, this type of floppy disk device also has
It has become necessary to apply a tracking servo similar to that of a so-called hard disk.

Incidentally, for an 8-inch floppy disk device, a tracking method has been proposed in which positioning is performed using an optical interference grating. This method embeds servo information on the inner and outer circumference of the disk,
The radial deflection (eccentricity) of the inner and outer circumferences is measured in advance, and the track position is determined by interpolation. However, with this method, the servo information is only on the inner and outer peripheries of the disk, so it is not possible to accurately obtain track positions on the disk that expands and contracts anisotropically, and it is affected by mechanical positioning accuracy. It was difficult to obtain high track density.

On the other hand, apart from this, a so-called embedded servo system has been proposed. In this method, a data surface is divided into a plurality of sectors, a servo sector for controlling the head position is provided in an index portion for detecting each sector, and servo information is embedded in the servo sector in advance.

In this system, for example, as shown in FIG. 1, a servo sector 2 provided between data sectors 1 is constituted by an erase section 3, an AGC section 4, and a position section 5. The erase section 3 has a maximum erasing period in one track and is a section that detects the beginning of each sector. Further, the AGC section 4 is used for signals when adjusting signal levels between tracks and between sectors. The position section 5 is a section for detecting the position in the radial direction of the recording medium, and a two-phase servo signal 6, which is usually called a two-phase dipit pattern, is written therein. Note that writing by the head is prohibited in this servo sector 2.

For such a disk, if the head 8 is now on-tracking on the 4N+4 track as shown in FIG. 1, the signal detected on the servo sector 2 will be the signal shown in FIG. Here, assuming that the signals detected by the position section 5 are position signals A, B, C, and D in order, the position control of the head is performed as follows.

In other words, the position signals A, B, C, and D are the difference signals between AB and CD of each peak-held value, X=A-B Y=C-D, as a first signal group, and this first Let the sum signal and difference signal of the signal group U=X+Y V=X−Y be the second signal group. These signals X, Y,
The relationship between the size of U and V and the position of the head is the third
As shown in the figure.

When performing a seek operation, if the maximum movement speed of the head is less than 4 tracks per sector,
As is clear from FIG. 3, by determining the size of X and Y, the section L ₀ where the head exists,
L ₁ , L ₂ , and L ₃ can be determined, and by using U and Y, which show linear changes within the determined section, the accurate position of the head can be determined.

That is, to determine the sections L ₀ , L ₁ , L ₂ , and L ₃ , first obtain the signals X and Y, and L ₀ ; X0, Y>0 L ₁ ; X>0, Y0 L ₂ ; L ₃ ; By determining that X<0, Y0, the head existence section L ₀
~L ₃ can be known.

On the other hand, the position of the head within the track is
Using U and V, D ₀ =V/a+0.5; L ₀ D ₁ =0.5−U/a; L ₁ D ₂ =0.5−V/a; L ₂ D ₄ =U/a+0.5; L ₃ (where a is the amplitude value of X and Y shown in FIG. 3).

Thus, by storing in memory the section L _B in which the head was located in the previous sector and the position D _B within that section L _B , the section L _P in which the head is currently located and its section L can be stored in memory. The current position PP can be determined from the position D _P within _P. At the same time, the head movement speed v per sector can also be determined by v=PP-PB.

Therefore, by using such information, first, according to the preset speed table,
By roughly controlling the movement of the head through speed control and then, when the head approaches the target position, position control, it is possible to effectively perform the seek operation of the head. In other words, accurate head positioning is possible even when the disk itself expands and contracts anisotropically.

[Problems with conventional technology]

By the way, in such an embedded servo system, the track following performance improves as the number of servo sectors formed on the disk surface increases, but on the other hand, when the number of servo sectors is added, the area occupied by the data sector decreases. , the storage capacity is reduced. Conventionally, therefore, the number of disk divisions is generally kept to about 32, and 32 servo sectors are formed to achieve a balance between the two. However, in floppy disk devices, the stationary and thermal dimensional errors mentioned above are more pronounced than in hard disk devices, so the head positioning method using 32 servo sectors as described above does not allow the dimensional errors caused by the above-mentioned dimensional errors to occur. There was a problem that the head could not sufficiently follow the track deviation.

Furthermore, in a sample value control system, there is generally a phase delay due to sample and hold. This delay becomes more pronounced as the servo band approaches the sampling frequency, so the servo band is usually limited to about 1/7 of the sampling frequency. However,
In a floppy disk device, the disk rotation speed is 360 rpm and the number of sectors is about 32, so the sampling frequency is 192 Hz. Therefore, the servo band is about 27 Hz, and the gain of the servo system at 6 Hz, which is the fundamental frequency of track runout, is about 10-odd dB at most, resulting in the problem that sufficient servo is not applied.

[Purpose of the invention]

The present invention was made in view of the above problems, and its purpose is to secure a wide servo band without causing a decrease in data storage capacity, and to efficiently operate regardless of disk track fluctuation. Moreover, it is an object of the present invention to provide a head positioning method that can perform head positioning with high precision.

[Summary of the invention]

The present invention divides a disk into a first area where data is recorded and a second area where data is not recorded, and uses servo information embedded in each area to position the head. The feature is that it is designed to do the following.

That is, in the second area, servo information (for example, servo information that can be considered continuous) that is more than twice the amount of servo information formed in the first area is written, and from this second area Based on the servo information read in advance, disk track runout information is obtained. When controlling the position of the head in the first area, the position control information calculated based on the servo information obtained from the first area and the track runout information are used to position the head. It is a feature.

〔Effect of the invention〕

According to the present invention, since the amount of servo information formed in the second area is larger than the amount of servo information formed in the first area, when the head runs within one sensor in the first area, Even if the head position information obtained in the first region is used, the head position information obtained in the first region can be interpolated with the track runout information obtained in the second region. As a result, head position control can be performed a plurality of times within one sector, and sufficient followability can be achieved for track shake within the sector.

Furthermore, by reading out the track shake information at regular intervals within one sector, the apparent sampling frequency can be improved. As a result, compared to the conventional embedding method, the servo band can be greatly expanded, and the gain of the servo system at the fundamental track vibration frequency can also be sufficiently improved.

Furthermore, the second area can be provided in an area where data is not normally recorded, and not only does this method cause almost no reduction in data storage capacity, but on the contrary, this method By using this, it is also possible to reduce the number of divided sectors in the first area and expand the data recording area.

Furthermore, in the present invention, by storing the track deflection information obtained from the second servo information, one head is shared for reading the first and second servo information. While the head positioning performance is greatly improved, the structure of the device can be made simple and compact, and the cost can be reduced.

[Embodiments of the invention]

Hereinafter, details of the present invention will be explained based on illustrated embodiments.

FIG. 4 schematically shows a recording pattern of a floppy disk according to an embodiment of the present invention. , a second region 13 is formed at the outer periphery. first area 12
is divided into 32 sectors, and each sector 14 is composed of a servo sector 15 and a data sector 16. That is, an embedded servo information pattern similar to the conventional one is formed in the first area.

On the other hand, the second area 13 is usually provided at the outermost circumferential portion where no data is written, and has a width of several tens of tracks. This area is divided into ten times the number of sectors as the first area, and each sector is a servo sector 17. Almost the same information as the servo sector 15 in the first area is written in this servo sector 17. As a result, servo information is almost continuously written into the second area.

The details of the servo sectors 15 and 17 are as shown in FIG. That is, the servo sectors 15 and 17 are connected to the erase section 18 and the AGC section 1.
9, a zone section 20, and a position section 21. The above zone part 20
is a portion provided to distinguish the first region and the second region, and the other portions have the same function as the conventional example. In addition,
Servo sectors 15 and 17 are prohibited from being written to by the head.

Next, a process in which track runout information is taken in from the disk 11 in which servo information is embedded as described above will be explained with reference to FIGS. 3 to 7.

In FIG. 6, when the disk 11 is inserted into the floppy disk device, the CPU 25 switches the servo system to speed control, and at the same time switches the analog switch 27 via the I/O port 26.
A negative speed control signal is given to the voice coil motor 31 via the I/O port 28, D/A converter 29, analog switch 27, and amplifier 30 (the direction of the current when going from the outer circumference to the inner circumference of the disk 11 is correct). As a result, the head 33 fixed to the carriage 32 is moved from the inner circumference to the outer circumference of the disk 11, and eventually the head 33 is fixed to the carriage 32.
contacts the elastic stopper 35. At this time, the head 33 is located in the second area 13 of the disk 11.
More specifically, it is set to be located at the center of the positive slope of the X signal shown in FIG. 3, that is, at the border between _L0 and _L3 . Further, the elastic force of the stopper 35 is set to such an extent that when the value of the current flowing through the voice coil motor 31 is changed, the stopper 35 can be moved to follow the eccentricity of the disk 11, track vibration caused by errors in temperature and humidity, etc. ing.

Thus, the head 33 is fixed at this position, reads servo information from 320 servo sectors 17 in one rotation of the disk 11, obtains the X signal in each sector, and stores it in the memory 36. The process at this time will be explained based on FIGS. 5 and 6.

When the head 33 is now in the position shown in FIG. 5, the head 33 first passes through the erase section 18 of the servo sector 17. Erase detection circuit 37
is composed of a retriggerable mono-multivibrator, and changes its output state after a certain period of time from the last fall of the signal recorded at the end of the previous sector. This generates a reference pulse for sampling. Next, head 33
AGC section 19 is detected. AGC signal is amplifier 3
8. The signal is sent to the AGC voltage generation circuit 41 via the analog switch 39 and the AGC circuit 40, and the signal level is adjusted. On the other hand, the sampling pulse generation circuit 42 sends a gate pulse delayed for a certain period from the reference pulse to the zone detection circuit 43, and the zone detection circuit 43 checks the presence or absence of a signal detected by the head 33 in the zone section 20. The CPU 25 determines whether the head is in the first or second area via the I/O port 26.
tell. The CPU 25 prepares for sampling corresponding to this.

When the head 33 reaches the position section 21, the aforementioned position signals A, B, C, and D are sampled and held. The sampling timing is
Sampling pulse generation circuit 4 from CPU25
The peak hold circuits 45, 46, 47, and 48 sequentially hold each peak value of the signals A, B, C, and D, respectively. Thus, the differential amplifiers 49 and 50 have X,
Y signals are output respectively. After passing through an analog switch 51 and an A/D converter 52, the X and Y signals are taken into the CPU via an I/O port 28.

Through the above process, when the X and Y signals are taken into the CPU 25, the CPU 25 transfers the X signal to the memory 3 while confirming that the Y signal is a constant value.
6. The capacity of the memory 36 is 360 bytes if one sample value is stored in one byte.

The X signal stored in the memory 36 is used as track runout information. The total sum of this track runout information becomes zero when the head 33 is at an average position with respect to track runout as shown in FIG. 7P. In addition, if the deviation is in the positive direction, the sum is a positive number, and if the deviation is in the negative direction, the sum is a negative number.If the absolute value exceeds a certain value, As shown in R, part of the track runout information will be missing, which is not preferable. Therefore, the CPU 25 calculates the average value from the track runout information, and determines whether the average value is within a predetermined range (for example, ±a/10). If the head position deviates from this range, the current value applied to the voice coil motor 31 is adjusted so that the head position falls within the above range, and track runout information is obtained again. .

Note that the track deviation information obtained is not necessarily
Since the average value is not necessarily zero, the average value is subtracted from the track runout information, and each piece of track runout information is corrected so that the offset value becomes zero.

The thus obtained track deflection information is used for head positioning when the head 33 reads/writes data in the first area 12.

Next, when the head 33 performs a seek operation to the target track in the first region, it first seeks according to a preset speed table until it comes within a predetermined distance, for example, 0.3 tracks, from the target track. , the speed of the heads 33 is controlled. Head position detection at this time is performed according to the same algorithm as the conventional method described above.

On the other hand, when the head 33 moves to the target track position and shifts to position control, the CPU 2 in FIG.
Step 5 determines which pattern from 4N to 4N+3 the target track includes, and selects the output of the differential amplifier 49 in accordance with the track as follows.

X; 4N track - Y; 4N+1 track - It is for the purpose of doing.

For example, if the target track consists of a 4N track pattern, the X signal is selected as the output of the differential amplifier 49. This X signal is sent to the analog switch 51, A/D converter 52, I/O
The data is taken into the CPU 25 via the port 28.
The CPU 25 controls the current of the voice coil motor 31 so that this X signal becomes zero. moreover
The CPU 25 reads from the memory 36 the track deflection information obtained from the servo sector 17 corresponding to the same position as the servo sector 15 from which the X signal was taken, and calculates the difference between this information and the X signal. This difference is used as a new offset value for track runout information until the head 33 detects the next servo sector 15. That is, the CPU 25 is
10 while the head 33 passes through one sector 14.
Track runout information is read out sequentially at regular intervals,
After correcting the track shake information using the offset value, the current of the voice coil motor 31 is controlled via the I/O port 28, D/A converter 29, compensation circuit 53, analog switch 27, and amplifier 30. As a result, the position of the head 33 is controlled ten times while it passes through one sector 14. Note that the compensation circuit 53 is a phase lead circuit,
It consists of an LPF, etc., and has functions such as compensating for phase lag due to sample and hold and limiting the servo band.

As described above, according to this embodiment, while the head 33 passes through one sector in the first area, ten pieces of track shake information are used to perform ten position corrections. Therefore, if the rotational speed of the disk is now 360 rpm, the number of disk divisions is 32, so the apparent sampling frequency in the first region has increased from the conventional 192 Hz.
This means that the frequency has been increased to 1920Hz, which is 10 times higher. As a result, the servo band can be increased to 270Hz, which is 10 times the conventional frequency, and the gain of the servo system is expected to be improved by more than 40 dB compared to the conventional system at around 6Hz, which is the fundamental frequency of track runout. in this way,
In the method of this embodiment, it is possible to sufficiently track not only the fundamental frequency of track runout but also its harmonic components.

In this embodiment, the X signal read by the fixed head 33 was used as the track deflection information, but conversely, this X signal was used to track the head 33, and another position detector, For example, the degree of deflection of the head 33 may be detected using a beam light emitter and a two-split photodetector.

In this case, for example, as shown in FIG.
A beam-shaped light emitter 55 is provided in a part of the carriage 32, and a two-split photodetector 56 is provided at a fixed position exactly facing the beam-shaped light emitter 55 when the carriage 32 comes into contact with the stopper 35. At this time,
Adjustment is performed by the voice coil motor 31 so that the head 33 is located at the midpoint between L ₀ and L ₃ in the second region, that is, near the midpoint of the positive slope of the X signal.

In this state, accurate tracking of the head 33 is performed using the X signal. The deflection signal q of the head 33 at this time is determined from the two outputs α and β of the two-split photodetector 56 as follows.

q=γα−β/α+β (γ is a proportional constant) This vibration signal q is output from the position detection circuit 57, and is sent to the analog switch 58, A/D converter 52,
is taken into the CPU 25 via the I/O port 28,
This information is stored in the memory 36 as track deviation information. Note that this information is further offset-corrected, as described above.

According to the method of this embodiment, the position of the head 33 is accurately controlled based on 320 pieces of servo information in the second region, so that the relative position of the beam-shaped light emitter 55 and the two-split photodetector 56 is controlled. If it is accurate and the linearity of the two-split photodetector 56 is high, it becomes possible to immediately obtain track runout information.

Note that the acquisition of track runout information in the second area is not limited to only after the power is turned on to the device. For example, it may be performed frequently after the power is turned on, or at longer intervals as time passes, using the standby time of the device. Make sure to do it accordingly. In this way, it is possible to sufficiently cope with expansion and contraction of the disk over time due to temperature and humidity.

It goes without saying that the accuracy of track runout information can be further improved by using the average value after multiple acquisitions, or by using the average value of information obtained from multiple tracks. do not have.

Further, if the track runout exceeds ±1/2 track, by switching between the X signal and the Y signal, it is possible to cope with a larger track runout.

Further, in the above embodiment, an elastic stopper is used for positioning the head when track runout information is taken in, but it is also conceivable to use a piezoelectric element to finely adjust the head position.

On the other hand, the read timing of the track deviation information during head positioning in the first region may be determined by multiplying the reference pulse generated by the erase detection circuit by 10 using a frequency multiplier circuit and applying this to the CPU as a clock signal.

In the above embodiment, 10 times as much servo information as in the first area was embedded in the second area, but the present invention can fully exhibit its effects by embedding at least twice as much servo information as in the first area. It is possible. and,
It goes without saying that the present invention is not limited to floppy disks, but can also be applied to hard disks, optical disks, etc. depending on the application.

[Brief explanation of the drawing]

Fig. 1 is a diagram for explaining the servo sector recording pattern of the conventional embedded servo system, Fig. 2 is a waveform diagram showing an example of a signal read by the head on the recording pattern, and Fig. 3 is a diagram showing the X, Y, V , a relationship diagram showing the relationship between the U signal and the head position, FIG. 4 is a diagram schematically showing the recording pattern of a floppy disk according to an embodiment of the present invention, and FIG. 5 is a diagram showing the servo sector portion of the recording pattern. 6 is a block diagram showing the electrical configuration of the floppy disk device according to the same embodiment, FIG. 7 is a relationship diagram showing the relationship between track runout information and offset value, and FIG. 8 is a diagram showing the relationship between track runout information and offset value. FIG. 2 is a block diagram showing the electrical configuration of a floppy disk device according to another embodiment of the invention. 1, 16...data sector, 2, 15, 17...
... Servo sector, 3, 18 ... Erase section, 4,
19...AGC section, 5,21...Position section,
8, 33...head, 11...floppy disk, 12...first area, 13...second area,
14...Sector, 20...Zone part, 35...Stopper, 55...Beam-shaped light emitter, 56...2
Split photodetector.

Claims (1)

[Claims] 1. First servo information is embedded in each of a plurality of servo sectors provided between a plurality of data sectors obtained by dividing a data recording area of a disk recording medium, and a first servo information consisting of these data sectors and servo sectors is formed. Second servo information, which is larger than the first servo information, is embedded in a second area on the disk recording medium excluding the first area, and is obtained by reading the second servo information in advance by a head. Track deflection information of the disk recording medium is stored, and positioning control of the head with respect to the target track is performed in accordance with the track deflection information and the first servo information read by the head. head positioning method. 2. The head positioning method according to claim 1, wherein the second servo information is embedded and formed substantially continuously in a second area set on the outer periphery of the disk recording medium. 3. The head positioning system according to claim 1, wherein the second servo information has an information amount n times that of the second servo information (n is an integer of 2 or more). 4. The head positioning system according to claim 1, wherein the first and second servo information are composed of two-phase dive patterns. 5 The track runout information obtained from the second servo information is obtained from the track runout information of the head whose tracking is controlled in accordance with the second servo information obtained when the head position is fixedly determined. A head positioning system according to claim 1.