JP3207839B2 - Method of controlling magnetic disk drive and magnetic disk drive - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a magnetic disk drive control method and a magnetic disk drive for performing high-density recording on a magnetic disk.

2. Description of the Related Art In response to the recent trend toward smaller and larger-capacity magnetic disk devices, the recording density of magnetic disks has been improved, that is, the track density and bit density have been improved. In order to improve the bit density, a magnetoresistive head (MR head) capable of increasing the level of a reproduction signal is used as a read head instead of a conventional inductive head capable of reading / writing. For this reason, there are two magnetic heads, an inductive head for writing and an MR head for reading.

Since the positions of these two heads in the track direction are different from each other, it is particularly desired to improve the read characteristics of the MR head.

[0004]

2. Description of the Related Art FIG. 24 is an explanatory diagram of an MR head, and FIG. 25 is an explanatory diagram of a yaw angle.

To improve the bit density, a magnetic head 9
The MR head 90-2 is used as the 0 read head. As shown in FIG. 24, the read head 90-2 includes:
When this MR head is used, it is necessary to separately provide the write head 90-1. For example, an inductive head is used for light. The use of such an MR head causes various problems to be improved.

First, since the MR head 90-2 is a magnetoresistive element, it reads out data by applying a bias current and utilizing a change in resistance value according to a magnetic force. This M
The R head 90-2 has a portion in which a change in resistance to a magnetic force is not a linear characteristic. Therefore, the bias current is adjusted, and the operating point is set to a portion where the change in resistance with respect to the magnetic force is linear. Conventionally, this adjustment was manually performed.

Second, the MR head 90-2 reads out data by supplying a bias current as described above. Therefore, a voltage is applied to the MR head 90-2. Conventionally, a voltage is applied to the MR head 90-2 simultaneously with turning on the power.

[0008]

However, the prior art has the following problems.

In the conventional manual adjustment of the bias current, it is difficult to set the optimum bias current because the manual adjustment is performed while viewing the reproduced waveform with an oscilloscope or the like. For this reason, the read margin has been reduced.

Conventionally, the MR head 90
Since the voltage is applied to −2, the voltage is applied before the magnetic disk does not rotate sufficiently. For this reason, the MR head 9
Since a voltage is applied during the floating of 0-2, there is a possibility that the MR head 90-2 may be damaged by a discharge with a magnetic disk having no potential. When a voltage is applied when it is not necessary, the characteristics of the MR head 90-2 are deteriorated due to a change with time.

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a method and an apparatus for controlling a magnetic disk drive for performing a read operation favorably by a read head separated from a write head.

It is another object of the present invention to provide a magnetic disk drive control method and apparatus for optimally setting a bias current of a read head separated from a write head.

Another object of the present invention is to provide a magnetic disk drive control method and apparatus for protecting a read head separated from a write head.

[0014]

FIG. 1 is a diagram illustrating the principle of the present invention.

[0015] Claim 1 of the present invention, the plurality of magnetic heads <br/> having a magnetoresistive head for reading data of the write head and the magnetic disk for writing data on a magnetic disk, a plurality of magnetic heads In a control method for a magnetic disk drive having a rotary actuator positioned at a desired track position on a magnetic disk, a control circuit includes a magnetic resistor for each of the plurality of magnetic heads.
Measuring the optimum bias current value of the anti-head, a measuring step of storing in the memory, upon selection of the magnetoresistive head, corresponding to the selected magnetoresistive head of the memory
It reads out the optimum bias current value, the magnetoresistive heads
And supplying the same to the memory, wherein the measuring step comprises:
By changing the applied bias current value of the magnetoresistive head ,
Measuring the bias current value at which the read output of the magnetoresistive head is vertically symmetrical, and re-measuring the optimum bias value of the magnetoresistive head at regular intervals during use of the drive, thereby rewriting the memory . It is a step of updating.

[0016] Claim 2 of the present invention, a magnetic head having a magnetoresistive head for reading data of the write head and the magnetic disk for writing data on the magnetic disk, the magnetic head to a desired track position of the magnetic disk A control method for a magnetic disk drive having a rotary actuator for positioning, wherein a step of detecting whether a spindle motor rotating the magnetic disk from power-on has reached a steady-state rotation; and detecting and the step of initiating the supply of the bias current to said magnetoresistive head, the magnetoresistive heads
That the write head is writing data while the bias current is being supplied to the drive;
A step of detecting that there is no access for a while, and a step of detecting that there is no access during the data writing or for the predetermined time and stopping the current supply to the magnetoresistive head .

Claims [0017] The present invention 3, the plurality of magnetic heads <br/> having a magnetoresistive head for reading data of the write head and the magnetic disk for writing data on a magnetic disk, a plurality of magnetic heads A rotary actuator for positioning at a desired track position on a magnetic disk ;
The optimum bias current value of each of the plurality of magnetic heads is measured and stored in a memory.
When the magnetoresistive head selection is the selection of the memory
Read the optimum bias current value corresponding to the magnetoresistive head, the and a magnetoresistive head for supplying the control circuit, the control circuit changes the applied bias current value of the magnetoresistive head, the magnetoresistive measuring the bias current value read output of the head is vertically symmetrical, drive use
During use, the optimum bias of the magnetoresistive head is
The assembling value is measured again, and the memory is updated .

[0018] Claim 4 of the present invention, a magnetic head having a magnetoresistive head for reading data of the write head and the magnetic disk for writing data on the magnetic disk, the magnetic head to a desired track position of the magnetic disk A rotary actuator for positioning and a spindle motor for rotating the magnetic disk from turning on the power supply detects whether the rotation has reached a steady state, detects that the steady state has been reached, and supplies a bias current to the magnetoresistive head . A power supply control circuit for starting supply , wherein the power supply control circuit comprises:
During the supply of the bias current to the magnetoresistive head,
A current supply to the magnetoresistive head is stopped upon detecting that the write head is writing data or that the drive has not been accessed for a certain period of time.

[0019]

According to the first and third aspects of the present invention, the control circuit comprises a plurality of control circuits.
Automatically measure the bias current of each of the magnetoresistive heads ,
Stored in the memory, upon selection of the magnetoresistive head, corresponding to
A bias current value is supplied, and the bias current is automatically adjusted. For this reason, an appropriate bias current value for each magnetoresistive head can be set, and the trouble of adjustment can be omitted.

According to claims 2 and 4 of the present invention, the spindle
Since the power supply of the magnetoresistive head is not performed until the rotation of the magnetoresistive head reaches the steady rotation, discharge of the magnetoresistive head and the magnetic disk can be prevented, and the destruction of the magnetoresistive head can be prevented. Also, once the current supply is started, the data is being written.
When there is no access for a certain period of time , the current supply to the magnetoresistive head is stopped, so that deterioration of the magnetoresistive head can be prevented.

[0021]

(A) Description of yaw angle correction FIG. 2 is an overall block diagram of one embodiment of the present invention, and FIG.
FIG. 4 is a block diagram of the servo demodulation / VCM drive circuit of FIG. 3, and FIG. 5 is a block diagram of the data demodulation / encoding / decoding circuit of FIG.

As shown in FIG. 2, the magnetic disk drive 1 includes a magnetic disk 100 rotated by a spindle motor 21, an actuator 101 rotating about a rotation shaft 102, and a VCM coil provided at the rear of the actuator 101. 20 and a magnetic head 105 provided at the tip of the actuator 101. The magnetic head 105 includes a write head 103 composed of an inductive element and a read head 1 composed of an MR element.
04.

The magnetic head 105 includes a head IC 23 for supplying a write current, a read bias current, and the like.
It is connected to the. The head IC 23 is connected to a read / write circuit 3b for demodulating read data and the like. The coil 20 of the actuator 20 is connected to a position circuit 3a for servo control. The hard disk controller 4 controls the read / write circuit 3b and the position circuit 3a, and
Of the desired track is read / written.

In FIG. 3, the servo head 10 reads servo information on the servo surface of the magnetic disk 100. The surf head 10 is also constituted by an MR head. The data read heads 11 to 1n are constituted by MR heads, and read data and servo information on the data surface of the magnetic disk 100. The VCM coil 20 forms a part of a VCM (voice coil motor) that moves the actuator 101. The spindle motor 21 rotates the magnetic disk 100.

The servo head IC 22 includes the servo head 1
0 is driven. The data head IC 23 selects and drives a write head and read heads 11 to 1n (not shown) according to a head select signal. The digital signal processor 30 calculates the amount of movement in response to the seek instruction, servo-controls the VCM coil 20, and controls the spindle motor 21 at a constant speed. The memory 300 stores data necessary for controlling the digital signal processor 30.

The servo demodulation circuit 31
And demodulates the servo signal read from the digital signal processor and outputs the signal to the digital signal processor 30. The VCM driver circuit 32 drives the VCM coil 20 according to the command amount of the digital signal processor 30. The DCM driver circuit 33 drives the spindle motor 33 according to the command amount of the digital signal processor 30.

The communication IC 36 mediates between the digital signal processor 30 and an interface processor 40 described later. These constitute the position circuit 3a.

The demodulation circuit 34 demodulates read data from the data head IC, outputs a Viterbi code as read data, and outputs a cylinder servo signal to the digital signal processor 30. The encoding / decoding circuit 35 decodes the Viterbi code from the demodulation circuit 34 and outputs the Viterbi code to the hard disk control circuit 42, encodes the write data into the Viterbi code, and outputs it to the data head IC 23.

The bias current control circuit 38 controls the data head IC 2
3 to control the bias current flowing to the MR heads 11 to 1n. The read / write circuit 3b includes the demodulation circuit 34, the encoding / decoding circuit 35, and the bias current control circuit 38.

Interface control processor 40
(Hereinafter, referred to as a processor) performs various interface controls. The memory 41 stores various data and the like necessary for the processing of the processor 40. The hard disk control circuit 42 has a SCSI (small computer interface) -2 interface, and performs analysis of a received signal and control of data transmission and reception.

The buffer 43 stores data necessary for processing of the hard disk control circuit 42. The timer 44 performs time lapse necessary for the processing of the processor 40. The servo error detection circuit 45 detects a servo error by monitoring the interval between markers.

The hard disk controller 4 is constituted by these components.

As shown in FIG. 4, the servo demodulation circuit 31
Is the servo head 1 via the servo head IC 31-1.
A servo demodulation circuit 31-1 for demodulating a servo signal of 0;
An AD converter 31 that converts the demodulated signal from analog to digital and outputs it to a digital signal processor 30
And 2. The VCM drive circuit 32 is a digital
It has a DA converter 32-1 for converting a drive signal from the signal processor 30 to digital / analog conversion, and a power amplifier 32-2 for driving the VCM coil 20 by the output of the DA converter 32-1.

As shown in the block diagram of the data demodulation / encoding / decoding circuit in FIG. 5, the AGC amplifier 340 controls the read signals of the read heads 11 to 1n by AGC. The filter circuit 341 cuts a high frequency component of the output of the AGC amplifier 340. The equalizer circuit 343 automatically equalizes the output of the filter circuit 341. The maximum likelihood detection circuit 344 detects a Viterbi code. VFO circuit 345
Generates a synchronous clock signal from the output of the equalizer circuit 343.

The data surface servo amplifier 346 reads the read heads 11 to 1n to calculate the off-track correction amount.
Amplifies the read servo signal recorded on the data surface of the magnetic disk. The peak hold circuit 347 detects the peak of the output of the servo amplifier 346. The AD (analog / digital) converter 348 converts the peak hold output of the peak hold circuit 347 into a digital value and inputs the digital value to the digital signal processor 30.

These constitute a demodulation circuit 34.
The 8/9 encoding / decoding circuit 35 encodes the 8-bit write data into 9-bit data, outputs the data to the write head, and converts the 9-bit read data into 8-bit NRZ data. Output as read data.

FIG. 6 is a configuration diagram of the yaw angle offset table in the configuration of FIG. The yaw angle offset tables 300-1 to 300-n are stored in the memory 300 of the digital signal processor 30. The yaw angle offset tables 300-1 to 300-n are provided corresponding to the read heads (MR heads) 11 to 1n, respectively. This yaw angle offset table 300-1 to 300-3
Each of 00-n is constituted by a correspondence table between a cylinder address and its yaw angle offset amount.

This yaw angle offset amount is set for every 128 cylinders. For example, 0-1 of the read head 11
Up to 27 cylinders, the yaw angle offset amount is y10. These settings are set by calibration
Automatically measured. That is, by changing the yaw angle offset amount,
Performs seek and automatically measures the yaw angle offset amount that maximizes the read margin, and sets the yaw angle offset table 3
00-1 to 300-n.

FIG. 7 is a flowchart of the read seek process, FIG. 8 is a flowchart of the write seek process, and FIGS. 9A and 9B are explanatory diagrams of the operation.

The read seek operation will be described with reference to FIG. (S1) As shown in FIG. 7, the hard disk control circuit 4
2. When the processor 40 receives the read seek instruction from 2,
The processor 40 calculates a target cylinder from the target read block. And the processor 40 is a digital
It instructs the signal processor 30 to perform a read seek to the target cylinder. (S2) The digital signal processor 30 refers to the yaw angle offset tables 300-1 to 300-n based on the target cylinder and the head selection signal and refers to the yaw angle offset amount (yaw) corresponding to the selected head and target cylinder. Angle correction amount).

Then, the digital signal processor 3
0 calculates the cylinder movement amount from the target cylinder and the current cylinder. (S3) On the other hand, the memory 300 of the digital signal processor 30 stores the amount of off-track at each track position. Therefore, the digital signal processor 30 reads the off-track correction amount of the cylinder position on the magnetic disk surface corresponding to the selected head. Then, the digital signal processor 30 calculates the final movement amount by adding the cylinder movement amount, the yaw angle offset amount, and the off-track correction amount.

Then, the digital signal processor 3
0 is the VCM driver circuit 32 according to the final movement amount.
The VCM coil 20 is driven via the. As a result, the actuator 101 seeks toward the target track. (S4) When the above-described seek of the movement amount is completed, the digital signal processor 30 notifies the processor 40 of the completion of the seek. As a result, the processor 40 outputs a head selection signal to the data head IC 23 and supplies a bias current to the corresponding head. At the same time, the hard disk control circuit 42 is notified. The hard disk control circuit 42 extracts a target read block from read data from the head via the demodulation circuit 34 and the encoding / decoding circuit 35, and sends the target read block to the SCSI interface.

Next, the write seek will be described with reference to FIG. (S1) As shown in FIG. 8, the hard disk control circuit 4
When processor 40 receives a write seek instruction from 2,
Processor 40 calculates a target cylinder from the target light block. And the processor 40 is a digital
Instruct the signal processor 30 to perform a write seek to the target cylinder. (S2) The digital signal processor 30 calculates a cylinder movement amount from the target cylinder and the current cylinder. (S3) The digital signal processor 30 reads the off-track correction amount of the cylinder position on the magnetic disk surface corresponding to the selected head. Next, digital
The signal processor 30 calculates the final movement amount by adding the cylinder movement amount and the off-track correction amount.

Then, the digital signal processor 3
0 is the VCM driver circuit 32 according to the final movement amount.
The VCM coil 20 is driven via the. As a result, the actuator 101 seeks toward the target track. (S4) When the above-described seek of the movement amount is completed, the digital signal processor 30 notifies the processor 40 of the completion of the seek. As a result, the processor 40 outputs a head selection signal to the data head IC 23, and applies a drive voltage to the corresponding write head. At the same time, the hard disk control circuit 42 is notified. The hard disk control circuit 42 supplies write data to the write head via the encoding / decoding circuit 35. As a result, the write head is driven, and write data is written to the target write block.

As shown in FIG. 9A, in the read seek, the seek movement amount is determined by the cylinder movement amount, the off-track correction amount, and the yaw angle offset amount. That is, since the yaw angle correction is performed, the core of the read head is located at the center of the write track on the inner side, the center side, and the outer side of the magnetic disk 100, and the read margin is increased.

On the other hand, as shown in FIG. 9B, in the write seek, the seek movement amount is determined by the cylinder movement amount and the off-track correction amount. That is, since the yaw angle correction is not performed, the core of the write head is located at the center of the write track on the inner side, the center side, and the outer side of the magnetic disk 100, and the write margin is widened.

At the time of this write, especially at the time of format write and data write, prior to the write, the ID
Need to lead the department. At this time, as shown in FIG. 9B, since the core of the read head is not located at the center, there is a possibility of a read error. However, the ID portion to be read has a short length, is fixed in length unlike the data portion, and has a lower update frequency than the data portion, so that the probability of occurrence of a read error is small.

Therefore, at the time of writing, there is little need to perform yaw angle correction in reading the ID section. Further, when a read error occurs in reading the ID section, it can be remedied by retry waiting for one rotation. Considering the system throughput, this method has much higher data storage efficiency than the conventional technique in which the gap 2 between the ID section and the data section is lengthened.

In this example, the yaw angle is corrected at the time of reading, but the yaw angle is corrected at the time of writing.
The write head may be positioned at the center of the write track, and the read head may be positioned at the center of the write track without performing yaw angle correction during reading.

Next, regarding the calibration operation,
This will be described with reference to FIG.

The calibration operation includes an off-track amount detection operation for each head and a VCM magnetic force detection operation for each cylinder. In this embodiment, the calibration seek is not performed at once, but is performed in small steps. This makes it possible to execute the calibration seek while minimizing the decrease in the total throughput of the interface. (S1) The processor 40 includes the hard disk control circuit 4
It is determined whether a command has come from step 2. If it is determined that a command (a command such as read or write) has arrived, it instructs the digital signal processor 30 to execute the command. The digital signal processor 30 executes a command. For example, in the case of a read / write command, FIG.
And execute the command as shown in FIG.

After the execution of the command is completed, the command interval timer for measuring the command interval of the processor 40 is reset and started. (S2) If it is determined in step S1 that a command does not arrive or if the command is executed, the processor 40
It is determined whether the command interval timer has overflowed. (S3) The overflow of the command interval timer means that no command arrives for a certain period of time.
Therefore, when the command interval timer overflows, the digital signal processor 30 is instructed to execute a calibration seek that has not been executed among the divided calibration seeks a1 to aN. Thereby, the digital signal processor 30 executes the divided calibration seek. (S4) In step S2, when the command interval timer does not overflow or executes the divided calibration seek, the processor 40 checks whether the calibration seek timer for determining the calibration seek execution interval has overflowed. If the calibration seek timer does not overflow, the process returns to step S1 because the interval between executions of the calibration seek has not been reached. (S5) Conversely, if the calibration seek timer overflows, the interval of the execution of the calibration seek has been reached, so the processor 40 determines whether all the divided calibration seeks have been executed from its own management table. Find out. If the processor 40 determines that all of the divided calibration seeks have not been executed, the process returns to step S3. Conversely, if the processor 40 determines that all of the divided calibration seeks have already been performed, it resets and starts the calibration timer, and returns to step S1.

As described above, when the command does not arrive for a predetermined period while the calibration seek execution interval is not reached, the calibration seek is divided and executed, and when the calibration seek execution interval is reached, the remaining time is reached. To perform the calibration seek of
Improve the total throughput of the interface,
The calibration seek can be performed at regular intervals. (B) Description of bias current control FIG. 11 is an explanatory diagram of a bias current of an MR head, FIG. 12 is an explanatory diagram of an optimum bias current of an MR head, FIG. 13 is a block diagram for automatic adjustment of a bias current, and FIG. FIG.
FIG. 3 is a circuit diagram of a head IC of No. 3;

The resistivity of the MR head to the magnetic field is shown in FIG.
It changes as shown in FIG. Therefore, unless the operating point (bias current) is properly selected, the output voltages with respect to the input magnetic field do not match. In the example shown in the upper part of FIG. 11, since the operating point is low, the output voltage is vertically asymmetric. In the example shown in the lower part of FIG. 11, the output voltage is also vertically asymmetric because the operating point is high.

In the example of FIG. 12, since the operating point is appropriate, the output voltage is vertically symmetric. This operating point is determined by the bias current applied to the MR head. The conventional bias current determination method is to change the bias current supplied to the MR head, observe the output of the reproduced waveform for each head, determine the bias value at which the positive and negative pulse amplitude values are symmetrical, The value was set manually.

However, the above-described method relies on human visual observation, and poses problems of measurement complexity and errors.
In addition, time-consuming man-hours are required. Further, once the bias current value is fixed for each head,
Cannot follow changes due to temperature fluctuations. For this reason,
The error becomes large, which is a factor that lowers the reliability of data demodulation.

In this embodiment, the time required for manufacturing the apparatus is reduced by adding an automatic adjusting mechanism to the apparatus itself. In addition, by operating the automatic adjustment mechanism at certain time intervals during use of the apparatus, the apparatus can follow temperature fluctuations and aging, thereby significantly improving the reliability of the data demodulation function of the apparatus.

In FIG. 13, the same components as those shown in FIGS. 3 to 5 are denoted by the same symbols. As shown in FIG. 13, the demodulation circuit 34 is provided with a peak detection circuit 342 that detects the peak of the output of the filter circuit 341.

The bias current control circuit 38 includes a full-wave rectifier 380 for full-wave rectifying the output of the peak detector 342.
And an integrating circuit 381 for integrating the full-wave rectified output, and an AD converter 382 for converting the integrated output to a digital value and inputting the digital value to the processor 40. Further, it has a DA converter 383 that converts a bias current value from the processor 40 into an analog amount, and a bias circuit (bias amplifier) 384 that amplifies the output of the DA converter 383 and outputs the amplified output to the head IC 23.

Further, the processor 40 is provided with a memory 400 for storing the measured bias current values I1 to In for the head addresses 1 to n.

In the circuit diagram of the head IC shown in FIG. 14, the head driving circuit 23 corresponds to each of the MR heads 11 to 1n.
-1 to 23-n are provided. The head drive circuits 23-1 to 23-n are connected to the above-described bias circuit 384 via switches S1 to Sn which operate according to a head select signal.

The head drive circuits 23-1 to 23-n have the same configuration. Here, the head drive circuit 23-1
The circuit configuration will be described as an example. A reference voltage line including a series circuit from the power supply voltage VDD1 to the current source 231, the diode VF, the voltage source Vc, and the resistor Rc; and a power supply voltage VDD.
1 to a current source 232, a pair of resistors Ra, Ra, a current source 2
33, a reference current line composed of a series circuit reaching the ground. A capacitor C for holding a voltage is provided in parallel with the pair of resistors Ra.

Then, the difference between the reference voltage of the reference voltage line and the comparison voltage Vb of the resistance Ra of the reference current line is obtained by the comparison amplifier 230, and the current source 232 of the reference current line is controlled to obtain the current of the reference current line. Is kept at the reference value.

On the other hand, a series circuit from the power supply VDD2 to the first transistor VF1, the resistor Rb, the MR head 11, the resistor Rb, the current source 234, and the ground is formed. The base of the first transistor VF1 is connected between the current source 232 of the reference current line and the resistor Ra.
A second transistor VF2 having a collector connected to the power supply voltage VDD1, a base connected between the resistor Ra and the current source 233, and an emitter connected between the resistor Rb and the current source 234 is provided.

In this circuit, the constant current value of the current source 232 is controlled according to the bias current supplied from the bias circuit 384 via the switch S1. Thereby, MR
The current Is flowing through the head 11 is also proportional to the bias current value.

Next, the operation of automatically adjusting the bias current of the MR head will be described.

When the automatic adjustment operation is started, the processor 4
0 selects each head, outputs the initial bias current value to the DA converter 383, outputs the initial bias current value to the head IC 23 via the bias circuit 384, and allows the bias current corresponding to the MR head to flow.

Next, the read signal from the MR head is
The bias current control circuit 38 receives the signal through the C amplifier 340, the filter circuit 341, and the peak detection circuit 342. The bias current control circuit 38 performs full-wave rectification of the read signal detected by the full-wave rectifier circuit, and outputs the read signal to the integration circuit 3.
Integrate at 81. This integrated output is supplied to the AD converter 38
2 to the processor 40. If the integrated output is constant, the read waveform of the MR head is vertically symmetric. Accordingly, if the integrated output is not constant, the processor 40 changes the bias current value and measures the integrated output of the read waveform.

In this way, the processor 40 measures the optimum bias current value at which the integrated output is constant or almost the most constant, and stores it in the head address column of the memory 400. Then, the processor 40 measures the optimum bias current value for all the MR heads and stores it in the memory 400.

As described above, in the adjustment process at the time of shipment from the factory, the bias current is automatically adjusted. Then, during normal use, the corresponding optimum bias current value of the memory 400 is read from the head address selected by the processor 40. Then, the processor 40 outputs the bias current value to the DA converter 383, and the bias circuit 3
The bias current is output to the head IC 23 via 84 and a bias current corresponding to the MR head flows. Thereby, each MR
Since the output of the head is vertically symmetrical, a read error can be prevented.

The optimum bias current differs depending on the temperature change and the aging. For this reason, during use of the apparatus, the processor 40 executes the automatic adjustment at regular intervals and updates the optimum bias current value of the memory 400 when no command has arrived. When this is performed at the time of the calibration seek described with reference to FIG. 9, the efficiency is high.

As described above, since the optimum bias current is automatically adjusted, the measurement is easy and the error is small.
In addition, the number of time steps can be reduced. (C) Description of Power Control FIG. 15 is a block diagram of the power control, FIG. 16 is a circuit diagram of the power control circuit of FIG. 15, and FIG. 17 is a circuit diagram of the bias current control circuit of FIG.

In the conventional magnetic disk drive, the power is supplied to the data writing circuit, the data demodulation circuit, and the head IC at the same time when the power is turned on. Also, during the period from when the power is turned on until the spindle motor reaches a steady rotation, during data writing, or when the device is not accessed for a long time, the read operation is not performed. In some cases, a bias current is supplied to the MR head.

On the other hand, since the data read / write operation is not performed before the spindle motor reaches the steady rotation, supplying power during that time causes unnecessary power consumption. Further, since the head flies before the spindle motor reaches a steady rotation, a potential difference occurs between the MR head and the magnetic disk. This gives M
Electric current or discharge occurs between the R head and the magnetic disk, and the head life is shortened due to deterioration of the MR head.

Further, if the bias current is continuously supplied unnecessarily, the head life is shortened due to the deterioration of the MR head, and the output characteristics of the MR head are deteriorated.

In this embodiment, after the spindle motor has reached a steady rotation, power is supplied to prevent such a problem.

In FIG. 15, a power supply 370 supplies power. The power supply control circuit 371 includes a head IC 23 from the power supply 370, a data writing circuit (encoding / decoding circuit).
35, controlling the power supply of the data demodulation circuit 34; The power supply control circuit 371 will be described later with reference to FIG. The steady-state rotation detection circuit 372 measures the time interval of the hall signal generated every one rotation of the spindle motor 21 to detect whether the spindle motor 21 has reached the steady-state rotation.

The bias current control circuit 38-1 outputs the steady-state rotation detection output of the steady-state rotation detection circuit 372 and the processor 40
The supply of the bias current of the MR head to the head IC 23 is controlled by the write gate signal and the no access signal for a predetermined time. The bias current control circuit 38-1 is configured as shown in FIG.
7 will be described later.

As shown in FIG. 16, power supply control circuit 371
A drive circuit 371-1 driven by a steady-state rotation detection signal of a steady-state rotation detection circuit 372 which becomes a low level after a steady-state rotation is detected, a coil driven by an output of the drive circuit 371-1, a power supply 370 and a data writing circuit 35 , A data demodulation circuit 34 and a relay 371-2 for connecting to the head IC 23.

As shown in FIG. 17, the bias current control circuit 38-1 includes a NOR circuit 385 and an inverter circuit 38.
6 and a resistor 387. The NOR circuit 385 includes a steady-state rotation detection circuit 372 that goes low after the steady-state rotation is detected.
, A write signal that goes high during data writing, and a no-access signal that goes high when there is no access for a fixed time, and outputs the inverted output of the OR. The resistor 387 is the resistor Rc of the head drive circuit 23-1 of the head IC 23 shown in FIG.

This operation will be described. Steady-state rotation detection circuit 3
Numeral 72 measures the time interval of the hall signal of the spindle motor 21. If the time interval between the hall signals is not constant, a high-level steady-state rotation detection signal is output as unsteady rotation. Conversely, if the time interval of the hall signal is constant, a low-level steady-state rotation detection signal is output as the steady-state rotation.

As shown in FIG. 16, when power is supplied to the magnetic disk device and the spindle motor 21 is in an irregular rotation until the spindle motor 21 reaches a steady rotation, a current flows through the coil of the relay 371-2 of the power supply control circuit 371. Absent. Therefore, the switch of the relay 371-2 is in the off state, and the data write circuit 35, the data demodulation circuit 34, the head IC 2
No power is supplied to 3.

On the other hand, when the spindle motor 21 reaches the steady rotation, the relay 37 of the power supply control circuit 371
A current flows through the coil 1-2. Therefore, the relay 37
The switch 1-2 is turned on, and power is supplied to the data write circuit 35, the data demodulation circuit 34, and the head IC 23.

As a result, unnecessary power is not supplied to the data write circuit 35 and the data demodulation circuit 34, thereby saving power. In addition, since unnecessary power is not supplied to the head IC 23, no voltage is supplied to the MR head until the spindle motor rotates normally. For this reason, no voltage is supplied to the MR head while the head is floating, so that energization or discharge between the MR head and the magnetic disk can be prevented. Thereby, deterioration of the MR head can be prevented.

Next, the processor 40 sets the write gate signal to high level during data writing. Conversely, when data is not being written, the write gate signal is set to low level. Further, the processor 40 sets the no access signal to a high level when there is no access for a predetermined time. Conversely, otherwise, the no-access signal is set to low level.

As shown in FIG. 17, if the steady rotation detection signal is at a high level, the write gate signal is at a high level, or the no access signal is at a high level, no current flows through the resistor 387. Therefore, no current flows through the MR head. That is, no current flows through the MR head when the spindle motor 21 has not reached steady rotation, during data writing, or when there has been no access for a certain period of time.

On the other hand, when the steady rotation detection signal is at a low level,
When the write gate signal is at a low level and the no access signal satisfies all the conditions of a low level, a current flows through the resistor 387 and a current flows through the MR head. That is, when the spindle motor 21 has reached a steady rotation and data is not being written and access is not performed for a certain period of time, current flows to the MR head.

Therefore, no bias current is supplied to the MR head until the spindle motor rotates normally. For this reason, no bias current is supplied to the MR head while the head is flying, so that energization or discharge between the MR head and the magnetic disk can be prevented. Thereby, deterioration of the MR head can be prevented. (D) Description of Servo Error Detection FIG. 18 is a block diagram of a servo error detection circuit, and FIG. 19 is a circuit diagram of a main part of FIG.

A method is known in which a marker is provided at the head of a servo pattern on the servo surface of a magnetic disk. For example, a track having 215 servo patterns on one track has 215 markers. in this way,
The servo control system performs processing immediately after the marker detection circuit detects a marker from the servo information.

However, when an abnormality such as the information stored on the servo surface is destroyed for some reason, an erroneous servo process is performed by the abnormal servo information. For this reason, a situation occurs in which data is destroyed, and a failure occurs in that the control system stops.

In this embodiment, if the servo information is normal, the markers appear at regular intervals. Therefore, the destruction of the servo information is detected by detecting the markers and observing the intervals at which the markers appear. is there.

In FIG. 18, the count data load timing generation circuit 450 outputs a marker detection signal MKR.
-FND changes from low level to high level, so that the counter save signal CNT-SAVE changes from low level to high level and the counter load signal CNT-LOAD changes from high level to low in synchronization with the PLL clock PLL-CLK. Level, PLL clock PLL-C
It is changed by one period of LK.

The count data load timing generation circuit 450 will be described later in detail with reference to FIG. Note that the marker is detected by a marker detection circuit (not shown) provided in the communication IC 36 in FIG. or,
The PLL clock CLK is synchronized with the information written on the servo surface.

The counter 451 outputs the counter load signal C
In response to NT-LOAD, an initial value ("0") is loaded, and the PLL clock CLK-CLK is counted. The counter data storage circuit 452 stores the counter save signal CNT.
-Store the count value of the counter 451 according to SAVE.

In the comparison data storage circuit 453, the marker interval is stored in advance by the processor 40 as comparison data. The comparison circuit 454 includes a counter data storage circuit 452
Is compared with the storage data of the comparison data storage circuit 453, and outputs a servo error signal ERR-SV to the processor 40 if they do not match. The details of the comparison circuit 454 will be described later with reference to FIG.

As shown in FIG. 19, the count data load timing generation circuit 450
A first flip-flop 450-1 for setting a marker detection signal MKR-FND by L-CLK, and a PLL
A second flip-flop 450-2 that sets the output of the first flip-flop 450-1 by the clock PLL-CLK, and a third that sets the output of the second flip-flop 450-2 by the PLL clock CLK-CLK Flip-flop 450-3, first flip-flop 450-1 and second flip-flop 450-2
And the counter save signal CNT-SAV
E, an AND gate 450-5, a second flip-flop 450-2, and a third flip-flop 450
-3 and counter load signal CNT-L
As shown in FIG. 19, the comparison circuit 454 includes a NAND gate 450-4 for outputting OAD, and the count data CNT-DAT of the counter data storage circuit 452.
A and a 16-bit EOR circuit 454-1 for performing an exclusive OR operation on the comparison data CMP-DATA of the comparison data storage circuit 453, and an OR circuit 454 for performing an OR operation on the 16-bit data of the EOR circuit 454-1 -2 and a flip-flop circuit 4 for latching the output of the OR circuit 454-2
54-3.

The operation will be described. A marker detection circuit detects the marker and outputs a marker detection signal MKR-FND
Changes to a high level, the PLL clock PLL-C
In synchronization with LK, the counter save signal CNT-LOAD
Changes to the high level, and the count value of the counter 451 is set in the counter data storage circuit 452. next,
One cycle later than the PLL clock PLL-CLK, the counter load signal CNT-LOAD goes low. As a result, the counter 451 is loaded with the initial value “0”, and counts the PLL clock PLL-CLK from the initial value.

In the comparison circuit 454, the count data of the counter data storage circuit 452 and the comparison data storage circuit 4
Each bit is compared with 53 comparison data. Then, when the comparison circuit 454 detects a comparison mismatch, the comparison circuit 454 outputs a servo error signal ERR-SV.

The processor 40 monitors the servo error signal ERR-SV from the comparison circuit 454. If the servo error signal ERR-SV falls,
It is determined that a servo error has occurred. As a result, servo error processing is performed. For example, stop servo control,
Take measures such as reporting to a higher rank.

As described above, using the markers provided at regular intervals on the servo surface, the appearance intervals of the markers are observed, and abnormalities in the servo surface information are detected. Therefore, a servo error can be detected quickly. Thereby, malfunction of the device can be prevented at an early stage. (E) Description of Timer Circuit FIG. 20 is a block diagram of the timer circuit, FIG. 21 is a circuit diagram of the start / stop control circuit of FIG. 20, FIG. 22 is a circuit diagram of the operation mode switching circuit of FIG. 20, and FIG. 3 is a time chart of the circuit of FIG.

Usually, a timer circuit for measuring a specific time is used for the processor 40. In a conventional timer circuit, after setting the count value in a register,
Start was specified in the start / stop bit.
In this case, at least two processes are required to start the timer.

Further, the timer does not have a function of detecting an interrupt signal, and even if the interrupt signal is asserted, the timer does not stop and the timer times out.

For this reason, at least two processes are required to start the timer, which imposes a burden on the processor 40 that controls the entire timer. Thus, even if the timer is configured by hardware, it takes time to operate the timer.

Similarly, even if an interrupt signal is detected, there is no function to stop the timer. Therefore, even if the interrupt signal is asserted, the timer times out and the interrupt signal is notified to the processor 40. For this reason, it is necessary for the processor 40 to process the interrupt signal from the timer, which increases the load on the processor 40.

This embodiment is a timer circuit for realizing a timer function with a minimum burden on a processor.

As shown in FIG. 20, the timer 44 has a register 440, an interrupt signal detection circuit 441, a start / stop control circuit 442, an operation mode switching circuit 443, and a counter circuit 444 having a prescaler 445. .

The register 440 sets a count value, an operation mode switching bit, and a start / stop bit from the processor 40. Interrupt detection circuit 441
Detects the interrupt signal and instructs the start / stop control circuit 442 to stop the timer. The start / stop control circuit 442 includes the register 4
The start / stop of the timer is controlled by the start / stop bit set to 40.

The operation mode switching circuit 443 controls the switching of the operation mode of the timer according to the operation mode switching bit set in the register 440. The counter circuit 444 reads the count value set in the register 440, decrements the value, and returns the value to the register. The prescaler 445 divides the frequency of the external clock CLK to generate various clocks CLK and CLK.
1, CLK2 and CLK3 are generated.

The start / stop control circuit 442 will be described with reference to FIG. The start / stop bit flip-flop 440-1 of the register 440 receives a start / stop instruction from the processor 40.
Set by a write signal from NOR circuit 44
2-1 outputs a load clock LOADCLK in synchronization with a write signal when a start / stop bit is set in the flip-flop 440-1.

The signal generation circuit 442-2 outputs the clock CL
A first flip-flop 442a in which the Q output of flip-flop 440-1 is set by K, and a clock C
It has an inverter circuit 442c for inverting LK, and a second flip-flop 442b in which the Q output of the first flip-flop 442a is set by the clock CLK of the inverter circuit 442c. In this circuit, a clock stop signal CLKSTOP is output from the inverted Q output of the first flip-flop 442a. The load / start signal LOAD / START output from the second flip-flop 442b is output in synchronization with the clock CLK and delayed from the write signal.

The NOR circuit 442-3 is connected to the processor 40
And the NOR of the interrupt signal from the interrupt detection circuit 441 and the two flip-flops 442a and 442b of the register 440 and the signal generation circuit 442-2.

Next, the operation mode switching circuit 443 will be described with reference to FIGS. As shown in FIG. 22, a clock selection circuit 443-1 selects a group of gates 443a to 443a to select one of the three clocks CLK1, CLK2, and CLK3 divided by the prescaler 445 according to the two operation mode bits. 443d. This output is as shown by X in FIG.

The synchronizing circuit 443-2 includes a flip-flop 443e for setting the operation clock selected by the clock selecting circuit 443-1 by the clock CLK, the above-described clock stop signal CLKSTOP, the output of the flip-flop 443e, and the clock CLK. And a NOR circuit 443f for taking NOR. In this circuit, an operation clock (indicated by Y in FIG. 23) synchronized with the clock is output, and the output of the operation clock is stopped when the processor 40 specifies the stop.

The NOR circuit 443-3 includes a synchronous circuit 443
-2 operation clock and load clock signal LOAD
It takes NOR with CLK and outputs a clock signal CLKOUT (shown in FIG. 23) of the counter 444.

The operation will be described. The processor 40
To operate the timer, the start / stop bit is designated as start, and the register 440 (flip-flop 4
When writing to 40-1), the load clock signal LOADCLK is output from the NOR circuit 442-1 to the operation mode switching circuit 443. In the operation mode switching circuit 443, a clock signal for the counter 444 is generated as described above. The signal generation circuit 442-
In 2, the load / start signal LOAD / START is generated after the write signal to the register 440.

For this reason, the load / start signal LOAD
While / START is at a low level, the load clock signal LOADCLK is input to the counter 444 as the clock signal CLKOUT. The counter 444 enters the load state when the load / start signal LOAD / START is at a low level and the clock signal CLKOUT is at a high level. Therefore, the counter 444 is loaded with the count value of the register 440. Next, load /
When the start signal LOAD / START becomes high level, the counter 444 enters a countable state. Therefore, the counter 444 outputs the clock signal CLKOUT
Count down.

Thus, even if the count value and the start designation are set simultaneously from the processor 40, the timer can be operated.

Next, the processor 40 sets a count value, an operation mode, and a start / stop bit to switch the operation mode of the timer. In addition to the operation described above, the clock selection circuit 443-1 includes the register 4
The clock used in the counter 444 is selected from the 40 operation mode switching bits, and the clock signal CLKOU
Output as T. Therefore, even if the count value, the start designation, and the operation mode setting bit are simultaneously set from the processor 40, the timer can be operated.

Next, when the processor 40 designates stop, the flip-flop 442a of the start / stop control circuit 442 is reset. As a result, a high-level clock stop signal CLKSTOP is output.
The output of 43-2 maintains a high level. For this reason,
The supply of the clock to the counter 444 is stopped.

Therefore, even if the stop is designated by the processor 40, the timer can be stopped while leaving the count value up to the present.

Similarly, when an interrupt signal or a reset signal arrives, the start / stop control circuit
The flip-flop 442a is reset by the R circuit 442-3. Accordingly, the high-level clock stop signal CLKSTOP is output, so that the output of the synchronization circuit 443-2 of the operation mode switching circuit 443 maintains the high level. Therefore, the supply of the clock to the counter 444 is stopped.

Therefore, even when an interrupt signal arrives, the timer can be stopped while leaving the count value up to the present. Thereby, the load on the processor 40 due to timeout can be reduced. The reset signal is
Since it is also provided to the counter 444, when the reset signal arrives, the counter 444 is also reset.

Thus, the processor can start or stop the timer with a single instruction to load the timer value and generate a signal to start the timer inside the timer. This greatly contributes to reducing the load on the processor. (F) Description of Other Embodiments In addition to the above-described embodiments, the present invention can be modified as follows.

In the embodiments shown in FIGS. 7 and 9, the read head is an MR element and the write head is an inductive element. However, the present invention can be applied to other read heads and write heads.

The present invention has been described with reference to the embodiments.
Various modifications are possible within the scope of the present invention, and these are not excluded from the scope of the present invention.

[0126]

As described above, according to the present invention,
The following effects are obtained.

According to claims 1 and 3 of the present invention, the control circuit comprises:
Automatically measures the bias current of each of multiple magnetoresistive heads
And store them in memory.
Since a corresponding bias current value is supplied and the bias current is automatically adjusted, an appropriate bias current value for each magneto- resistive head can be set, and the labor for adjustment can be omitted .

According to claims 2 and 4 of the present invention, the spindle
Until the motor reaches steady rotation,
Since no supply is performed, the magnetoresistive head and magnetic disk
To prevent discharge of the magnetic head and the destruction of the magnetoresistive head.
You. Also, once the current supply is started, the data is being written.
Or if there is no access for a certain period of time,
Stopping the flow supply prevents deterioration of the magnetoresistive head

[Brief description of the drawings]

FIG. 1 is a principle diagram of the present invention.

FIG. 2 is an overall block diagram of an embodiment of the present invention.

FIG. 3 is a detailed block diagram of the circuit of FIG. 2;

FIG. 4 is a block diagram of a servo demodulation / VCM drive circuit of the circuit of FIG. 3;

FIG. 5 is a block diagram of a data demodulation / encoding / decoding circuit of the circuit of FIG. 3;

FIG. 6 is a configuration diagram of a yaw angle offset table in the circuit of FIG. 3;

FIG. 7 is a flowchart of a read seek process according to an embodiment of the present invention.

FIG. 8 is a flowchart of a write seek process according to an embodiment of the present invention.

FIG. 9 is an operation explanatory diagram of FIGS. 7 and 8;

FIG. 10 is a flowchart of a calibration process according to an embodiment of the present invention.

FIG. 11 is an explanatory diagram of a bias current of the MR head.

FIG. 12 is an explanatory diagram of an optimum bias current of the MR head.

FIG. 13 is a block diagram for automatically adjusting the optimum bias current of the MR head.

FIG. 14 is a circuit diagram of the head IC in FIG.

FIG. 15 is a block diagram for power supply control.

16 is a circuit diagram of the power supply control circuit of FIG.

17 is a circuit diagram of the bias current control circuit shown in FIG.

FIG. 18 is a block diagram of a servo error detection circuit.

19 is a main part circuit diagram of FIG. 19;

FIG. 20 is a block diagram of a timer circuit.

21 is a circuit diagram of the start / stop control circuit of FIG.

FIG. 22 is a circuit diagram of the operation mode switching circuit of FIG. 20;

FIG. 23 is a time chart of the circuit of FIG. 22;

FIG. 24 is an explanatory diagram of an MR head.

FIG. 25 is an explanatory diagram of a yaw angle.

FIG. 26 is an explanatory diagram of a conventional data format.

[Explanation of symbols]

3a read / write circuit 3b positioning circuit 4 hard disk controller 10-1n read head (MR head) 20 VCM (coil) 23 head IC 30 digital signal processor (first control circuit) 34 demodulation circuit (read circuit) 35 encoding / Decoding circuit (write circuit) 38 Bias current control circuit 40 Processor (second control circuit) 44 Timer circuit 45 Servo error detection circuit

──────────────────────────────────────────────────続 き Continuing on the front page (72) Yoshiyuki Nagasaka 4-1-1, Kamidadanaka, Nakahara-ku, Kawasaki-shi, Kanagawa Prefecture Inside Fujitsu Limited (72) Tamotsu Kozuka 4-1-1 Kamiodanaka, Nakahara-ku, Kawasaki-shi, Kanagawa Prefecture No. 1 Inside Fujitsu Limited (72) Inventor Susumu Yoshida 4-1-1 Kamikadanaka Nakahara-ku, Kawasaki City, Kanagawa Prefecture Inside Fujitsu Limited (56) References JP-A-5-182113 (JP, A) JP-A-5-6504 (JP, A) JP-A-5-342747 (JP, A) JP-A-4-364266 (JP, A) JP-A-1-51711 (JP, A) JP-A-1-42069 (JP) JP-A-62-204613 (JP, A) JP-A-56-35116 (JP, U) (58) Fields studied (Int. Cl. ⁷ , DB name) G11B 5/02 G11B 5/39

Claims (4)

(57) [Claims] 1. A light for writing data to the magnetic disk
Magnetoresistive head and reads data of the magnetic disk
A plurality of magnetic heads having a head, a control method for a magnetic disk drive having a rotary actuator for positioning the plurality of magnetic heads to a desired track position of the magnetic disk, the control circuit, the plurality of To each magnetic resistance of the magnetic head
Measuring the optimum bias current value of the head, a measuring step of storing in the memory, upon selection of the magnetoresistive head, the selection of the memory
Reading the optimum bias current value corresponding to the performed magnetoresistive head, and supplying the optimum bias current value to the magnetoresistive head. The measuring step includes changing an applied bias current value of the magnetoresistive head ,
A step of measuring the bias current value read output of the magnetoresistive head is vertically symmetrical, in the drive use, at regular time intervals, said magnetoresistive f
And re-measuring the optimum bias value of the head, the control method of a magnetic disk drive which is a step of updating the memory. 2. Write for writing data to a magnetic disk
Magnetoresistive head and reads data of the magnetic disk
A magnetic head having a head, a control method for a magnetic disk drive having a rotary actuator for positioning the magnetic head to a desired track position of the magnetic disk, spindle motor for rotating the magnetic disk from power
A step of chromatography data to detect whether the reached steady rotation, it is detected that reaches the steady rotation, the magnetoresistive f
A step of starting the supply of bias current to the head, while the supply of the bias current to said magnetoresistive head, the
The write head is writing data , and
A step of detecting that there is no access to the drive for a certain time; and a step of detecting that there is no access during the data writing or the certain time and stopping the current supply to the magnetoresistive head . A method for controlling a magnetic disk drive, comprising: 3. A write write data to the magnetic disk
Magnetoresistive head and reads data of the magnetic disk
Measurement and a plurality of magnetic heads, a rotary actuator for positioning the plurality of magnetic heads to a desired track position of the magnetic disk, the optimum bias current value of the magnetoresistive head of each of said plurality of magnetic heads having a head And store it in memory
When the magnetoresistive head selection is the selection of the memory
A control circuit for reading an optimal bias current value corresponding to the magnetoresistive head, and supplying the optimum bias current value to the magnetoresistive head , wherein the control circuit changes an applied bias current value of the magnetoresistive head ,
A bias current value at which the read output of the magnetoresistive head is vertically symmetrical is measured, and the drive is used at regular intervals.
Then, the optimum bias value of the magnetoresistive head was measured again.
And updating the memory . 4. A write write data to the magnetic disk
Magnetoresistive head and reads data of the magnetic disk
A magnetic head having a head, and a rotary actuator for positioning the magnetic head to a desired track position of the magnetic disk, spindle motor for rotating the magnetic disk from power
Over data detects whether reached steady rotation, it is detected that reaches the steady rotation, and a power control circuit for starting the supply of the bias current to said magnetoresistive head, the power supply control circuit During supply of a bias current to the magnetoresistive head,
A magnetic disk drive, wherein the current supply to the magnetoresistive head is stopped upon detecting that the write head is writing data or that the drive has not been accessed for a certain period of time.