JP3718158B2 - Disk unit - Google Patents

Description

[0001]
[Industrial application fields]
The present invention relates to an improved small-sized disk device that performs position sensitivity measurement processing and generates sector pulses in consideration of encoding loss at the time of reading and decoding loss at the time of reading.
[0002]
[Prior art]
In recent years, the magnetic disk apparatus has been packed with a smaller track pitch as the size and capacity thereof are increased, and an improvement in on-track accuracy is required. In addition, since the MR head with a narrow core width is adopted as the read head for the write head using the inductive head, it is essential to correct the core deviation between the write head and the read head. The switching position may be an on-track position.
[0003]
For this reason, two-phase servo information for changing the phase at the 1/3 track pitch is used instead of the conventional one in which the phase of the two-phase servo pattern is changed at the 1/2 track pitch.
[0004]
[Problems to be solved by the invention]
However, the position sensitivity correction value in the conventional disk device may be measured by measuring the cross points of the two-phase servo signals N and Q. However, the servo signals N and Q are obtained discretely at predetermined sample periods. The cross point cannot be measured directly. Therefore, an error occurs because the value of the crossing point is obtained by linear interpolation from the values before and after crossing the crossing point of the two-phase servo signals N and Q obtained at the constant speed seek.
[0005]
Also, in the case of a two-phase servo signal that changes the phase at a 1/3 track pitch by closing the track pitch, there are two cross points during the movement of one track, compared to the case of a 1/2 track pitch. Since the density of the points has doubled, it takes too much time to measure the position sensitivity correction value, and there is a problem that improvement in measurement accuracy cannot be expected due to linear interpolation.
[0006]
Furthermore, in order to reduce the size and increase the capacity of disk devices, the size of the disk has been reduced, and the capacity has been increased by high-density recording of the disk. Higher functions are being promoted by adopting partial response maximum likelihood (PRML) or the like.
[0007]
As the signal processing system becomes highly functional, a time loss in encoding and decoding, which has not been a problem with the conventional 1-7 RLL, has increased. For example, in 1-7RLL, the loss was only about 5 bits, but in the signal processing by the partial response maximum likelihood method, for example, there are about 10 times as much as 44 bits. There is a problem that the format efficiency decreases.
[0008]
In particular, in a format that takes into account the loss of encoding and decoding in the past, an encoding loss that extends from the end of inputting NRZ data during a write operation to the end of writing to a disk medium, and a read signal during a read operation Therefore, there is a problem that the format efficiency is lowered because a gap area corresponding to the loss including both the decoding loss until the NRZ data is actually demodulated and output is provided.
[0009]
An object of the present invention is to provide a disk device that improves the on-track accuracy by improving the measurement time and accuracy of the position sensitivity correction value even if the track pitch of the two-phase servo information is clogged.
[0011]
[Means for solving problems]
FIG. 1 is a diagram illustrating the principle of the present invention.
[0012]
The present invention converts a two-phase servo signal N, Q detected from a read signal of two-phase servo information recorded on the disk surface into correct head position information using a position sensitivity correction value measured in advance. Intended for equipment.
[0013]
With respect to such a disk device, the present invention is arranged so that the phase changes at a 1/3 cylinder pitch on the disk surface of the disk medium as shown in FIG. In addition, two-phase servo signals N and Q cross points are generated in the order of 2 places, 1 place, 2 places and 1 place. Two-phase servo information is recorded, and when the position sensitivity correction value is measured by the position sensitivity measurement unit, on-track control is performed on the cross points 102 and 104 of the two-phase servo signals N and Q. A value of 104 is measured to obtain a position sensitivity correction value.
[0014]
Here, the position sensitivity measurement unit obtains the measurement value at the position 102 where the cross point is on the plus side and the measurement value at the position 104 where the cross point two tracks away is minus, and averages the absolute values of the two measurement values. The value is the position sensitivity correction value. Thereby, the vertical asymmetry of the read waveform by the MR head can be relaxed.
[0015]
The position sensitivity measurement unit calculates a correction coefficient for correcting the measured value of the cross point to the theoretical value as the position sensitivity correction value.
[0016]
Further, the position sensitivity measuring unit divides the recording area of the disk medium into a plurality of zones, measures and stores the position sensitivity correction value for each zone boundary position, and controls the zone in which the head is located during head position control. The position sensitivity correction value at the current position is calculated by linear interpolation from the position sensitivity correction values at the two zone boundary positions.
[0021]
DETAILED DESCRIPTION OF THE INVENTION
<Contents>
1. Overall structure
2. FPC mounting structure
3. Store offset correction data in RAM
4). Measurement of position sensitivity correction value
5. Encoding / decoding loss
1. Overall structure
FIG. 2 is a block diagram of the overall circuit configuration of the disk apparatus of the present invention. A disk drive 10 that is a disk device of the present invention includes a control unit 12 and an enclosure 14. The enclosure 14 is provided with a disk 18 as a storage medium rotated by a spindle motor 16.
[0022]
In the present invention, for example, two 2.5-inch magnetic disks are used. The enclosure 14 is provided with a voice coil motor (hereinafter referred to as “VCM”) 20 for driving the head actuator so that the head 22 can be positioned with respect to the data surface of the disk 18.
[0023]
In this embodiment, since there are two disks 18, there are four data surfaces and four heads 22 are used. Further, the enclosure 14 is provided with a head IC 27, and performs read / write with respect to the control unit 12 with respect to the head 22, reading of servo information, head switching, and the like.
[0024]
The control unit 12 is provided with an MCU 24. The MCU 24 incorporates a ROM 62a and a RAM 64a as shown in the drawing in addition to the CPU. The ROM 62a and ROM 64a may be built in the MCU 24 or may be provided outside. For the MCU 24, an oscillator 26 that oscillates a predetermined clock, a logic IC 28 that generates a clock necessary for various controls based on the clock from the oscillator 26, a program memory 30 as an external ROM, and a spindle motor 16 of the enclosure 14 And a servo controller 32 for controlling the VCM 20, various commands necessary for input / output with the host system 40, a hard disk controller 34 for transferring data, a buffer memory 36, and a read / write for reading / writing to the disk 18. A light unit 38 is provided.
[0025]
In addition, in the disk drive 10 of the present invention, a shock sensor 41 is provided, and a detection signal from the shock sensor 41 is processed by the sensor IC 42 and applied to the MCU 24. For example, a piezoelectric element is used as the shock sensor 41. When the acceleration in a certain direction due to an external impact exceeds a specified value by the sensor IC 42, an impact detection signal is output to the MCU 24, and if a write operation to the disk 18 is in progress. This write operation is forcibly terminated.
[0026]
The piezoelectric element used as the shock sensor 41 has directionality. In this embodiment, the piezoelectric element is arranged so as to detect the impact of the head actuator in the direction of rotation of the head 12, that is, the direction across the track of the disk 18. .
[0027]
Furthermore, a console 43 can be provided outside the disk drive 10 as necessary, and input / output to / from the MCU 24 required for starting up the apparatus or for maintenance can be performed.
[0028]
FIG. 3 shows the internal structure of the enclosure 14 in the disk drive 10 of FIG. A disk 18 is provided on the base 21 of the enclosure by a spindle motor and is rotated at a predetermined speed. An actuator 44 is installed at the corner of the base 21 with respect to the disk 18, and the actuator 44 is rotated by the VCM 20 installed at the rear so that the head 22 at the tip can be positioned and moved in the radial direction of the disk 18.
[0029]
Near the actuator 44, a base-side FPC 48 on which the head IC 23 is mounted is installed. A read / write FPC band 46 is pulled out from the base side FPC 48 and supported and fixed to the side surface of the actuator 44 on the head 22 side.
[0030]
4 is a cross-sectional view taken along the line AA in FIG. An actuator 44 is rotatably mounted on the base 21 and the VCM 20 is installed at the rear. The tip of the actuator 44 is divided into three head arms, and the head 22 is installed at the tip of the head arm. Two disks 18 are connected to the rotating portion of the spindle motor 16, and four heads 22 are disposed relative to the data surfaces on both sides of the disk 18. A cover 23 is attached to the upper part of the base 21, and a printed circuit board 25 on which the control unit 12 of FIG.
2. FPC mounting structure
FIG. 5 shows the actuator 44 provided in the enclosure of FIG. 3 taken out together with the base side FPC 48. A read / write FPC band 46 is pulled out from the base side FPC 48, is sandwiched by the retainer 45 on the side surface of the rotation shaft of the actuator 44, and the tip portion taken out from the retainer 45 is connected to the head 22 at the FPC connection portion 50 of the head arm 60. It is connected by overlapping with the relay FPC installed on the side.
[0031]
FIG. 6 is a side view of the actuator 44 of FIG. 5 on the FPC connection portion 50 side, and the retainer 45 is fixed to the side face of the actuator 44 by screws 54. By fixing the retainer 45, the front end of the read / write FPC band 46 and the end of the relay FPC 52 mounted on the head 22 side are overlapped and fixed electrically and mechanically in the FPC connection section 50. .
[0032]
FIG. 7 is an enlarged view of the FPC connection unit 50 shown in FIG. The mounting surface of the FPC that is the side surface of the head arm 60 of the actuator 44 forms a stepped portion 56 that is lowered to the mounting side of the read / write FPC band 46. In the step portion 56, a tip end 46a of a read / write FPC band 46 having a tip attached to the retainer 45 is installed.
[0033]
In other words, the read / write FPC band 46 is formed on the side surface of the head arm 60 in a state where the front end portion 46a is attached to the retainer 45 with, for example, double-sided tape, after being inserted into the holding portion 45a of the retainer 45 at the front end side. Is mounted in the stepped portion 56.
[0034]
When the retainer 45 and the tip portion 46a of the read / write FPC band 46 are attached to the step portion 56, the surface of the read / write FPC band tip portion 46a, that is, the land portion where the connection pattern is formed, is fixed by the relay FPC 52. It is installed so as to be flush with the mounting surface. Connection and fixing of the relay FPC 52 to the read / write FPC band distal end portion 46 a bonded to the retainer 45 in the step portion 56 is performed by screwing the retainer 45 to the actuator 44 with a screw 54.
[0035]
FIG. 8 shows the relay FPC 52 that is bonded and fixed to the side surface of the actuator 44 shown in FIG. The relay FPC 52 is divided into three FPCs 52a, 52b, and 52c. Each of the FPCs 52a to 52c has land portions 58a, 58b, and 58c formed on the connection side of the relay FPC band 46, and land portions 59a to 59c formed on the head side. Each of the lands 58a to 58c and 59a to 59c is provided with a rectangular connection pattern.
[0036]
In this embodiment, the head 18 is a composite head, and includes a write head using an inductive head and a read head using an MR head. For this reason, in the land portions 59a to 59c in FIG. 8, four rectangular connection patterns are formed corresponding to the four sets of write head and read head, respectively. Also, four rectangular connection patterns are formed on each of the land portions 58a to 58c to which the read / write FPC bands are connected by overlapping corresponding to the four rectangular connection patterns.
[0037]
FIG. 9 shows the FPC mounting structure of the present invention shown in FIG. 7 in comparison with the conventional structure. FIG. 9A shows the FPC structure of the present invention, in which a stepped portion 56 is formed on the mounting surface of the head arm 60, and the retainer 45 and the land portion of the read-write FPC band 46 are placed in an overlapped state. Thus, the land side connection surface on the front side of the read / write FPC band 46 is positioned on the same plane as the mounting surface 52 a of the relay FPC 52 with respect to the head arm 60. For this reason, the connecting surfaces of the land portions at the ends of the read / write FPC band 46 and the relay FPC 52 can be made the same surface.
[0038]
On the other hand, in the conventional structure, as shown in FIGS. 9B, 9C, or 9D, the retainer 45 and the read-write FPC band 46 are superposed on the flat mounting surface of the head arm 60, and the tip thereof In addition, the land portions are overlapped and connected and fixed in a state where the tip of the relay FPC 52 is lifted from the opposite side.
[0039]
For this reason, a positional deviation in the left-right direction is generated by the amount of the land portion at the tip of the relay FPC 52 that is raised, and a deviation occurs between the designed land portion dimension and the actual positioning dimension, as shown in FIG. Such connection patterns of the land portions 58a to 58c are shifted from each other to cause a contact failure. Therefore, high-precision positioning is required to accurately align the rectangular connection patterns formed on the land portions. Such problems are all solved in the FPC structure of the present invention in FIG.
3. Store offset correction value in RAM
FIG. 10 is a functional block of storage and generation processing of offset correction data used for head position offset correction performed by the MCU 24 provided in the control unit 12 of FIG.
[0040]
In FIG. 10, a ROM table 62 and a RAM table 64 are used for storing offset correction data. The ROM table 62 and the RAM table 64 use the ROM element 62a and the RAM element 64a built in the MCU 24 of FIG.
[0041]
In the disk drive 10 of the present invention, a memory with a small memory capacity is used for the RAM built in the MCU 24 for cost reduction, and the area that can be used for storing offset correction data is limited. The offset correction data that effectively uses the limited RAM area is stored.
[0042]
FIG. 11 shows an embodiment of the ROM table 62 and the RAM table 64 by measuring the offset correction data for correcting the mechanical bias force with respect to the cylinder position of the disk medium by zone division.
[0043]
FIG. 11A shows the measured value of the offset current that flows through the VCM 20 to remove the mechanical external force on the two disks 18 shown in FIG. 4 with respect to the cylinder position. Since the disk 18 of the present invention has a storage area of, for example, 4096 cylinders, it is first divided into eight zones Z1 to Z8 in units of 512 cylinders.
[0044]
As shown in FIG. 3, this mechanical bias force depends on the bending force of the relay FPC band 46 that connects the base side FPC 48 to the actuator 44. For this reason, since the external force increases at both ends of the outer side and the inner side, a large offset current is required to correct this, and since the external force is stable in the central portion, the amount of change is relatively small. It is an offset current for external force correction that is almost linear.
[0045]
In FIG. 11A, when the external force offset characteristic 75A side is taken as an example, in the present invention, there is almost no change in the offset current in the central zones Z4 and Z5 among the four zones Z1 to Z8. Since it is a straight line, these two zones are regarded as one zone and the value of the offset correction current is stored. Here, the offset correction currents of the offset characteristics 75A measured for each of the zones Z1 to Z8 are I1 to I7.
[0046]
FIG. 11B shows a ROM table 62 having eight memory areas corresponding to the eight zones Z1 to Z8 of the cylinder position. Here, since the ROM element 62a built in the MCU 24 of FIG. 2 has a sufficient storage capacity compared to the RAM element 64a, there is no problem in capacity even if a memory area is provided for the zones Z1 to Z8.
[0047]
The eight memory areas of the ROM table 62 store pointer information P1 to P7 indicating the storage positions of the bias force correction currents I1 to I7 stored in the RAM table 64 shown in FIG. Since two zones Z4 and Z5 are regarded as one zone, the same pointer information P4 is stored, and the bias force correction current I4 in the fourth memory area of the RAM table 64 is referred to by the pointer information P4. I can do it.
[0048]
Thus, if it is desired to make a plurality of zones one zone according to the characteristics of the offset data in the zones Z1 to Z8 with respect to the cylinder direction, the same pointer information is stored in the ROM table 62 and the same in the ROM table 64. The offset correction data in the storage area may be referred to, and the number of memory areas in the RAM table 64 can be reduced by this amount.
[0049]
Referring to FIG. 10 again, an offset correction data generation unit 65 is provided for the ROM table 62 and the RAM table 64. In the offset correction data generation unit 65, the current cylinder position X detected from the head position signal is set in the register 66, and the zone address generation unit 68 first obtains the zone Zi to which the current cylinder position X belongs and sets it in the register 70. Then, the ROM table 62 is referred to.
[0050]
The ROM table 62 stores pointer information as shown in FIG. 11. The pointer information is read out, and the corresponding offset correction data in the ROM table 64 is read out. The offset correction data A that is read first is held in the register 74. Subsequently, a zone Zi + 1 next to the current zone Zi is generated from the zone address generation unit 68 and set in the register 72, and pointer information is obtained by referring to the ROM 62 by the zone Zi + 1. The corresponding offset correction data B is obtained with reference to the table 64 and stored in the register 76.
[0051]
That is, as apparent from FIG. 11A, for the zones Z1 to Z8, for example, the bias force correction current measured for the right zone boundary position is stored in the RAM 64 as the offset correction data I1 to I7.
[0052]
For example, if the head is present at the cylinder position X in the zone Z1, the pointer information P1 obtained by referring to the ROM table 62 by the zone Z1 indicates the zone boundary position on the left side of the zone Z1 by referring to the RAM table 64. Only the offset correction data I1 can be obtained.
[0053]
In the present invention, offset correction data at an arbitrary cylinder position in the zone is obtained by linear interpolation of the measured values at the zone boundaries on both sides, so the second ROM table 62 is referred to for the adjacent zone Z2. Thus, the pointer information P2 is obtained, the offset correction data I2 is obtained by referring to the RAM table 64, and the offset correction data I1 and I2 at the zone boundary positions on both sides of the zone Z1 are obtained for linear interpolation. This is the acquisition of the offset correction data A by the zone Z1 and the acquisition of the offset correction data B by the next zone Zi + 1 shown in FIG.
[0054]
If the offset correction data A and B are obtained at the zone boundary positions on both sides of the zone Zi to which the current cylinder position X belongs to the registers 74 and 76, the interpolation calculation unit 78 determines the interval C between the current cylinder position X and the zone Zi. Is used to calculate the offset correction data at the current cylinder position X as interpolation data. That is,
Interpolation data = A + [{(B−A) / C} × (X−A)]
Interpolation data can be calculated as
[0055]
In this way, the interpolation data generated by the offset correction data generation unit 65 is supplied, for example, by adding an offset correction current for removing the influence of the external force to the VCM to the current by the speed control at the time of seek control. Further, during on-track control, the obtained offset data is added to the position servo control loop based on the head position signal to remove the influence of the external force on the actuator.
[0056]
FIG. 12 is a flowchart of the processing operation by the offset correction data generation processing unit 65 of FIG. First, in step S1, the zone Zi is determined from the current cylinder position X, and pointer information Pi of the RAM table 64 is obtained by referring to the ROM table 62. Subsequently, in step S3, the ROM table 62 is referred to in the adjacent zone Zi + 1 to obtain the RAM pointer information Pi + 1.
[0057]
Subsequently, in step S4, the RAM table 64 is referred to by the pointer information Pi, Pi + 1, and correction values A and B are acquired in step S5. Subsequently, in step S6, linear interpolation calculation for obtaining a correction value for the current cylinder position is performed, and in step S7, the calculated interpolation data is output.
[0058]
FIG. 13 shows the zone division according to the embodiment of FIG. 10 and the contents of the ROM table 62 and the RAM table 64 for the position sensitivity correction data.
[0059]
FIG. 13A shows the position sensitivity correction value for the cylinder position. The position sensitivity correction value is a coefficient for converting the head position read from the servo information of the disk medium into a theoretically determined correct head position, and this is used as the position sensitivity correction value K. The position sensitivity correction value K changes relatively slowly as shown in FIG. 13A, for example.
[0060]
Therefore, in FIG. 13A, as in the case of the external force correction in FIG. 11, the cylinder direction is divided into eight zones Z1 to Z8 at equal intervals, and as shown in FIG. A ROM table 62 having eight memory areas is prepared. Here, the characteristics of the position sensitivity correction value in the cylinder direction increase linearly in the zones Z1 to Z5 and decrease linearly in the remaining zones Z6 to Z7.
[0061]
Here, the zone Z8 is an area that cannot be used by the user as a system zone, and the measured value of the zone boundary is fixedly used for the zone Z8. Therefore, the zones Z1 to Z5 are collectively regarded as one zone. Therefore, all pointer information P1 is stored in the area corresponding to the zones Z1 to Z5 of the ROM table in FIG. The pointer information P1 refers to the position sensitivity correction value K1 stored at the first position of the RAM table 64 in FIG.
[0062]
Next, the zones Z6 and Z7 are regarded as one zone, the same pointer information P2 is stored in the corresponding memory area of the ROM table 62, and the position measured at the zone boundary position on the left side of the zone Z of the RAM table 64 by the pointer information P2. The sensitivity correction value K2 is referred to. For the zone Z3, pointer information P3 is stored in the ROM table 62 so that the position sensitivity correction value K3 in the third memory area of the RAM table 64 can be referred to.
[0063]
For such a position sensitivity correction value that changes linearly, the straight line portion is regarded as one zone, and in this case, the memory area of the RAM table 64 is reduced to three for the eight zones Z1 to Z8. be able to.
[0064]
FIG. 14 shows an example of BL correction data for correcting fluctuations in the cylinder direction of BL (multiplier of magnetic flux B and coil length L) due to the permanent magnet used in the VCM. The contents of the RAM table 64 are shown.
[0065]
FIG. 14A shows the offset correction characteristic of the bias force with respect to the cylinder position, and the offset of the bias force is normally obtained as a measured value of the open loop gain G. The open loop gain G indicating the offset due to BL is small at the cylinder positions on the inner and outer sides, becomes flat at the center, and changes non-linearly during that time.
[0066]
As for the open loop gain for correcting BL in FIG. 14A, first, the memory area corresponding to the ROM table 62 is secured by dividing the zone into eight zones Z1 to Z8. Here, the measured values at the boundaries of Z1 to Z8 are G1, G2, G3, and G4, which are substantially symmetrical offset characteristics.
[0067]
Accordingly, the central zones Z4, Z5, and Z6 are regarded as one zone and the same pointer information P4 is stored in the ROM table 62 so that the open loop gain G4 of the RAM table 64 can be referred to. Further, the left and right zones Z3 and Z7 are regarded as symmetric, the same pointer information P3 is stored, and the open loop gain G3 of the RAM table 64 is referred to.
[0068]
Similarly, the zones Z2 and Z8 are regarded as symmetric zones, the same pointer information P2 is stored, and the open loop gain G2 of the RAM table 64 is referred to. Further, for the zone Z1, its own open loop gain G1 is stored in the RAM table 64.
[0069]
Further, with respect to the characteristics in the case of having a flat straight line portion in the center of FIG. 14, the interpolation calculation may not be performed for the zones Z4 and Z5. Interpolation calculation is performed only for the non-linear portions of zones Z1, Z2, Z3, Z6, and Z7. Further, since the zone Z8 is not a user area, for example, the open loop gain G1 is used in a fixed manner, and since it is a constant value, linear interpolation is not necessary even if this portion is reached.
[0070]
In the above embodiment, the external force offset correction data, the position sensitivity correction data, and the bias force correction data are taken as examples of the offset correction data. However, other than this, as appropriate according to the cylinder position used in the disk device. This offset correction data can be applied as it is.
4). Measurement of position sensitivity correction value
FIG. 15 is a block diagram of the position sensitivity correction value measurement function in the disk apparatus of the present invention. In FIG. 15, the position sensitivity measurement unit 80 is activated at the time of shipment of the disk drive, at the time of startup, or at a desired calibration timing, etc., and read signals of the two-phase servo information recorded by dividing into sectors on the data surface of the disk. The position sensitivity correction value is measured based on the two-phase servo signals N and Q obtained from the above.
[0071]
The measurement of this position sensitivity correction value is as follows Do. First, the two-phase servo pattern on the magnetic disk used for measuring the position sensitivity correction value correction value of the present invention is as shown in FIG.
[0072]
FIG. 16A shows servo patterns for one servo frame for cylinder numbers 1 to 4 with the horizontal axis as the disk radial direction and the vertical axis as the disk rotation direction (track direction). For example, paying attention to cylinder number 1, the two-phase servo patterns are four patterns of A, B, C, and D, and the phases are switched by dividing the 1-track pitch into 1 / 3-track pitch.
[0073]
That is, pattern A is first recorded with a 1/3 track pitch width, then shifted with 1/3 track pitch, and pattern B is recorded with 2/3 pitch width, and then shifted with respect to pattern A by 1/3 track pitch. Pattern C is recorded with a 2/3 track pitch width, and pattern D is recorded with a 2/3 pitch width by shifting the pattern B by 1/3 track pitch.
[0074]
The last AGC pattern is continuously recorded in the disc radial direction. This is obtained when the MR head 18a as a read head seeks at a constant speed in the disk radial direction with respect to the two-phase servo patterns A to C whose phases change at 1 to 3 track pitches in FIG. The two-phase servo signals N and Q are as shown in FIG. Here, the servo signal N is the difference between the read signals of the patterns A and B, and the servo signal Q is the difference between the read signals of the patterns C and D. That is,
N = A−B
Q = CD
As obtained. Generation of the two-phase servo signals N and Q based on the read signals of the two-phase servo patterns A to C is performed by the servo signal generation unit 84 of FIG.
[0076]
Book In the measurement processing of the position sensitivity correction value of the invention, the head is on-track controlled so as to be on-tracked to the measurement positions 102 and 104 of the cylinder numbers 2 and 4 which are the cross points of the servo signals N and Q. A position sensitivity correction value is obtained from the signal value of the obtained cross point.
[0077]
Here, the measurement position 102 of the cross point of the cylinder number 2 is on the + side, the measurement position 104 of the cross point of the cylinder number 4 is on the − side, and each of the + cross point and the − cross point is measured. The error due to the vertical asymmetry in the read signal of the MR head 18a is relaxed.
[0078]
That is, the cross point measurement value is obtained from the average value of the absolute value of the measurement value of the + cross point obtained at the measurement position 102 and the measurement value of the − cross point obtained at the measurement position 104. The position sensitivity correction value is calculated by dividing the theoretical value of the cross point by this measured value.
[0079]
17 shows the head position signal generated from the two-phase servo signals N and Q and the two-phase servo signals N and Q for on-track control to the cross point of the measurement position by the position sensitivity measuring unit 80 in FIG. It is explanatory drawing of the signal which connected the position signal.
[0080]
FIG. 17A shows a two-phase servo obtained by a signal read by the MR head 18a from the two-phase servo patterns A to C that change phase at a one to two track pitch in FIG. 16A, which is the same as FIG. N, Q. These two-phase servo signals N and Q are converted into head position signals that change linearly from − to + as shown for cylinder numbers 1 and 3 in FIG. For the on-track position 100 as well, the characteristic line at that portion passes through the zero point.
[0081]
In actual head position control, as shown in FIG. 17 (C), each head position signal for one track pitch obtained in FIG. 17 (B) is continuously added by adding a predetermined offset. Process as a connected signal. However, with the head position signal at the normal on-track position 100, the measurement positions 102 and 104 where the two-phase servo signals N and Q cross like the measurement cylinders 2 and 4 cannot be controlled on track.
[0082]
Therefore, in the position sensitivity correction value measurement process of the present invention, for the measurement cylinders 2 and 4 where the cross points 102 and 104 exist, a conditional expression for obtaining the head position signal from the two-phase servo signals N and Q specific to the measurement cylinder. Set.
[0083]
FIG. 18 shows the conditions of the two-phase servo signals N and Q for generating the head position signal of FIG. 17B used in the measurement cylinders 2 and 4, and the formula for calculating the head position signal under each condition. 18 is applied to the measurement cylinders 2 and 4 in FIG. 17A, so that the head position signals are respectively obtained for the measurement positions 102 and 104 at the cross points as shown in FIG. 17B. Can be converted to a signal passing through the zero.
[0084]
Each signal curve for each 1/3 and rack pitch of FIG. 17B obtained by the measuring cylinders 2 and 4 is obtained including flat portions of the two-phase servo signals N and Q in FIG. Therefore, it is a broken line change in which the inclination is 45 ° at the zero-cross portion and the inclination is dull at both ends.
[0085]
Further, the head position signal obtained by the measuring cylinders 2 and 4 is used with an offset value according to the conditions of the two-phase servo signals N and Q determined in the first offset table 90 and the second offset table 92 shown in FIG. When connected together, the result is as shown in FIG.
[0086]
FIG. 17C shows a state in which the head position signal of FIG. 17B is connected to the track pitch of the measuring cylinder 2 using the first offset table 90 of FIG. By using the calculation formulas according to the conditions of FIG. 18 and the offset values according to FIG. 19 for the measurement cylinders 2 and 4 in this way, the head position signal is set to zero at the measured values 102 and 104 that become cross points. Produces a head position signal that can be controlled.
[0087]
The head position signal shown in FIGS. 17A, 17B, and 17C is generated by the head position generator 86 in FIG. 15, and the first offset table 90 of the measurement cylinder 2 is passed through the multiplexer 88. An offset value is used, and the offset value of the second offset table 92 is used for the measurement cylinder 4.
[0088]
The head position signal used in the offset measurement obtained by the head position generation unit 86 is given to the addition point 94, and the deviation from the target position signal indicating the measurement position of the cross point given by the position sensitivity measurement unit 80 is extracted. Then, this deviation is given to the position servo control unit 96, and current is supplied to the VCM from the addition point 98 to perform on-track control to the target position 102 or 104 serving as the cross point.
[0089]
Then, the position sensitivity correction value measurement unit 80 measures the value of the cross point from the values of the two-phase servo signals N and Q from the servo signal generation unit 84 obtained at that time in the on-track state to the zero cross point as the measurement position. Then, when the measured value of the + cross point and the measured value of the-cross point are obtained, the average value of the sum of the absolute values of both is calculated, and the theoretical value of the cross point is divided to correct the position sensitivity. Find the value. In the actual position sensitivity measurement, the average value of the measured values at a plurality of cross points obtained in the on-track state with the measuring cylinder is used.
[0090]
The position sensitivity measurement unit 80 instructs the seek control unit 82 to perform seek control to the measurement cylinder position. That is, in the position sensitivity measurement of the present invention, the cylinder position is divided into, for example, eight zones, and the position sensitivity correction value is measured for each cylinder position at the zone boundary. The zone boundary cylinder address is set in 82 to perform seek control, and after completion of seek control, the position servo control unit 96 performs on-track control to the measurement cylinder.
[0091]
Further, as shown in FIG. 17, when the measurement of the cross point on the + side at the first measurement position 102 is completed at the cylinder number 2, next, the seek control unit 82 is instructed to perform the 2-track seek, and the next measured value 104 After the seek to the measuring cylinder 4 having a head, a head position signal for on-tracking to the cross point at which the measured value 104 is obtained is generated and measurement processing is performed.
[0092]
FIG. 20 is a flowchart of the position sensitivity correction value measurement process in FIG. First, in step S1, the measurement zone is set, and then in step S2, the first measurement cylinder that becomes the boundary position of the measurement zone is set. In step S3, the measurement cylinder is sought, and in step S4, the on-track control is performed to the measurement position that is the cross point of the measurement cylinder.
[0093]
Subsequently, in step S6, +2 track seek is performed to the next measurement cylinder, and in step S7, the track is on-tracked to the position of the zero cross point of the seek measurement cylinder. In step S8, the voltage -Vc at the cross point is measured. Subsequently, in step S9, an average value of absolute values of the positive and negative measurement voltages + Vc and -Vc is obtained, and a position sensitivity correction value at the position is calculated by dividing a theoretical value of a predetermined cross point by the average value. And store it in a table. Subsequently, in step S10, it is checked whether or not all zones have ended, and the same processing is performed for the boundary cylinder positions of all zones.
[0094]
Thus, even if the servo pattern of the 2-phase servo signal is recorded with a phase change of 1/3 track pitch, the position of the cross point to be measured by the on-track control is Since it is the second and fourth tracks as described above, Since it exists every other track and the interval between the cross points is widened, the density to be measured at the cross points is reduced, and the measurement time can be shortened.
[0095]
In addition, since the value of the cross point is directly measured by on-tracking to the position that matches the cross point, the measurement accuracy of the zero cross point is different in each step compared to the case where the value before and after the conventional cross point is obtained by linear interpolation. The measurement accuracy of the position sensitivity correction value obtained from this measurement result can be greatly improved.
[0096]
In addition, since position sensitivity correction values are measured in zones, the number of measurement points does not increase as the number of cylinders increases. In addition, the capacity of the RAM table for storing the position sensitivity correction value can be reduced.
5. Encoding / decoding loss
FIG. 21 is a block diagram of a circuit unit for generating a write gate signal and a read gate signal used in the disk apparatus of the present invention.
[0097]
First, the disk medium of the present invention employs a data surface servo in which servo frames are recorded on a track at predetermined intervals, and the servo frames are divided into a plurality of sectors. Therefore, a sector mark is recorded at the head position of each sector provided after the servo frame, and a sector pulse can be obtained by reading this sector mark.
[0098]
In the disk apparatus of the present invention shown in the figure, in the read / write unit 38 of the control unit 12, signal processing according to the partial response maximum likelihood method, that is, encoding of NRZ data at the time of writing, and read operation The head read signal at that time is demodulated to NRZ, and a loss of, for example, about 40 bits occurs at each time of encoding and decoding.
[0099]
Conventionally, the processing shown in the time chart of FIG. 22A is performed for such encoding loss and decoding loss. FIG. 22A shows servo detection signals, sector pulses, write gate signals, and read gate signals. The servo frame signal is obtained by detecting the servo frame. The servo frame signal is divided into four sectors in this conventional example, and a sector mark is recorded at the head position of each sector, which is obtained as sector pulses indicated by numbers 1, 2, 3, and 4. It is done.
[0100]
The write gate signal is generated at the rise of the sector pulse following the servo frame signal during the write operation, and until the write NRZ signal is input to the read / write unit 38 in the write state of FIG. 2 for one sector. Is on. When the write gate signal is turned off, a gap region for the encode loss time Tw during the write operation is provided behind the write gate signal.
[0101]
On the other hand, the read gate signal generates a read gate signal for the sector pulse for the decoding loss time Tr from when the head read signal is obtained from the rise of the sector pulse during the read operation until the NRZ signal is output to the hard disk controller 34. Delayed and turned on for a read time of one sector.
[0102]
To refer to the write gate signal again, a gap region corresponding to the decode loss time Tr generated for the read gate signal is provided following the gap region corresponding to the encode loss time Tw after the write gate signal is turned off. .
[0103]
That is, in the conventional sector servo format, the gap corresponding to the loss time To (= Tw + Tr) obtained by adding the encode loss time Tw during the write operation and the decode loss time Tr during the read operation after the write gate signal is turned off. An area is provided. For this reason, the sector area must secure a gap area corresponding to the loss obtained by adding the encode loss and the decode loss, and the format efficiency is lowered.
[0104]
Therefore, in the present invention, as shown in the time chart of the servo frame, sector pulse, and write gate signals during the write operation in FIG. 22B, the write gate signal is turned on in the sector format of the disk medium. After that, the gap area after being turned off is only the encode loss time Tw in the write operation.
[0105]
For the sector format in which the gap area at the final position of each sector is the encoding loss time Tw at the time of write operation, the read operation is detected following the servo frame signal as shown in FIG. The sector pulse is delayed by the encode loss time Tr during the read operation, and a read gate signal is generated in synchronization with the rise of the delayed sector pulse.
[0106]
Due to a delay corresponding to the decode loss time Tr of the sector pulse during such a read operation, a gap region obtained by adding the decode loss time Tr to the encode loss time Tw is formed at the end position of the sector format as shown in FIG. Since there is no need to provide a gap area corresponding to the decoding loss time Tr, the format efficiency of the disk medium can be increased.
[0107]
For example, in the case of FIG. 22, the conventional servo frame has 4 sectors as shown in FIG. 22A, but in the format according to the present invention of FIG. The sector area can be increased.
[0108]
The operation of the circuit unit in FIG. 21 will be described. At the time of writing, the sector pulseless table 100 considering the encoding loss is loaded into the sector pulse generation circuit 102, and the sequencer 104 is activated by the sector pulse, and the write gate signal becomes active. Subsequently, when the buffer manager 106 recognizes that data has been sent from the buffer RAM 108 up to the gap area corresponding to the encoding loss, the sequencer 104 stops the write gate signal. This is repeated for each sector pulse.
[0109]
At the time of reading, the sector pulse table 100 considering the decoding loss is loaded into the sector pulse generation circuit 102, the sequencer 104 is activated by the sector pulse, and the read gate signal becomes active. Subsequently, when the buffer manager 106 recognizes that necessary data has been received in the buffer RAM 108, the sequencer 104 stops the read gate signal.
[0110]
Here, each sector pulse table 100 used at the time of writing and at the time of reading holds a value determined for each zone on the program, and sets a value corresponding to the zone at that time.
[0111]
21 is an example of a data surface servo in which servo frames are recorded on the data surface at regular intervals. However, a servo surface servo using a recording surface of dedicated servo information is also used in the sector format of the data surface. Can be applied in exactly the same way.
[0112]
【The invention's effect】
As described above, according to the present invention, the connection surfaces on which the connection patterns of both ends of the read / write FPC from the base side and the relay FPC on the head side in the head actuator are formed as the same mounting surface. Therefore, the designed positioning dimension and the actual positioning dimension can be made the same, and the positioning accuracy can be easily improved.
[0113]
In addition, since the read-write FPC band side is incorporated into the step portion formed by lowering the FPC mounting surface of the actuator together with the retainer, the read-write FPC can be stably attached to the head actuator.
Further, according to the present invention, the storage position of the RAM table storing the correction data can be arbitrarily designated by the pointer information of the ROM table, so that the correction interval can be varied and the use area of the RAM is minimized. Can be reduced.
[0114]
In addition, according to the present invention, even if the servo pattern of the two-phase servo signal is recorded with a phase change of 1/3 track pitch, the position of the cross point to be measured by on-track control exists every other track. As a result, the cross point interval is widened, thereby reducing the density of the cross point measurement target and reducing the measurement time.
[0115]
Furthermore, according to the present invention, the gap area provided at the final position of each sector on the track is a gap area corresponding to the encoding loss during the write operation, and the gap corresponding to the loss including both the conventional encoding loss and decoding loss is added. Compared with the case where the area is provided, the decrease in the format efficiency can be improved and the disk capacity can be increased.
[Brief description of the drawings]
FIG. 1 illustrates the principle of the present invention
FIG. 2 is a block diagram showing the overall configuration of the present invention.
3 is an explanatory diagram of the structure of the disk enclosure in FIG. 2;
4 is a cross-sectional view taken along the line AA in FIG.
FIG. 5 is an explanatory view of the actuator shown in FIG. 3 taken out together with the FPC.
6 is a side view of the actuator of FIG. 5 on the FPC mounting surface side.
7 is an enlarged view of the FPC mounting structure of FIG.
8 is an explanatory diagram of the relay FPC in FIG. 6;
FIG. 9 is an explanatory view showing the FPC mounting structure of the present invention in comparison with a conventional structure.
FIG. 10 is a functional block diagram of storage and interpolation generation of offset correction data according to the present invention.
11 is an explanatory diagram of the ROM table and the RAM table in FIG. 10 taking external force offset correction data as an example.
12 is a flowchart of processing for generating offset correction data in FIG. 10;
13 is an explanatory diagram of the ROM table and the RAM table in FIG. 10 taking position sensitivity correction data as an example.
14 is an explanatory diagram of the ROM table and the RAM table in FIG. 10 taking BL correction data as an example.
FIG. 15 is a functional block diagram of the Bojon sensitivity measurement process of the present invention.
FIG. 16 is an explanatory diagram of a two-phase servo pattern and a demodulated two-phase servo signal
FIG. 17 is an explanatory diagram of a signal obtained by connecting the two-phase servo signal, the head position signal, and the head position signal at the time of position sensitivity measurement in FIG.
18 is an explanatory diagram of a condition table used to calculate a head position signal at the measurement cylinder position at the cross point in FIG. 17;
FIG. 19 is an explanatory diagram of an offset table used for connecting the Hüne position signals at the cross points in FIG. 17;
FIG. 20 is a flowchart for measuring position sensitivity in FIG. 15;
FIG. 21 is a block diagram of a read gate and write gate generation unit used in the present invention.
FIG. 22 is a time chart showing generation of a read gate and a write gate according to FIG.
[Explanation of symbols]
10: Disk drive
12: Control unit
14: Enclosure
16: Spindle motor (SPM)
18: Disc
20: Voice coil motor (VCM)
22: Head
24: MCU (micro control unit)
26: Oscillator
28: Logic IC
30: Program memory
32: Servo controller
34: Hard disk controller (HDC)
36: Buffer memory
38: Read / write unit
40: Host system
42: Sensor IC
44: Actuator
46: FPC band for read / write
48: Base side FPC
50: FPC connection
52: Relay FPC
54: Screw
56: Step
58a-58c, 59a-59c: Land part
60: Head arm
62: ROM table
64: RAM table
66, 70, 72, 74, 76: Register
68: Zone address generator
78: Interpolation calculator
80: Position sensitivity measurement unit
82: Seek control unit
84: Servo signal generator
86: Head position generator
88: Multiplexer
90: First offset table
92: Second offset table
100: Sector pulse table
102: Sector pulse generation circuit
104: Sequencer
106: Buffer manager
108: Buffer RAM

Claims (2)

In a disk apparatus for converting two-phase servo signals N and Q detected from a read signal of two-phase servo information recorded on a disk surface into correct head position information using a position sensitivity correction value measured in advance. ,
In the state where the phase changes on the disk surface at a 1/3 cylinder pitch and the cross points of the two-phase servo signals N and Q are generated in the order of two places, one place, two places and one place. A disk medium on which servo information is recorded;
When measuring the position sensitivity correction value, the measured value at the position where the cross point in the second and fourth tracks of the two-phase servo signals N and Q obtained from the servo information is on the plus side and the cross point are A position sensitivity measurement unit that obtains a measurement value at a negative position and calculates the position sensitivity correction value as an average value of the absolute values of the two measurement values;
A disk device comprising:   2. The disk apparatus according to claim 1, wherein the position sensitivity measurement unit calculates a correction coefficient for correcting the measured value of the cross point to a theoretical value as a position sensitivity correction value.

Images