JPH03237663A - Inspection method for defect in recording medium of disk storage device - Google Patents

Abstract

PURPOSE:To rationally inspect a defect in the recording medium of a disk by means of a disk storage device itself by combining a step where inspection data is written in a state where a head is shifted from the specified position of respective tracks and a read/inspection step where inspection data is read and the defect is detected from the presence or absence of a read error. CONSTITUTION:Inspection data written into respective sectors in the write step 40 are all usual data that a disk storage device 80 reads and they are read as usual data in the read/inspection step 60. Then, the presence or absence of the defect D in the sector is inspected from the presence or absence of the read error at that time. Thus, the defect D can be detected at a standard corre sponding to performance which the disk storage device 80 has. Thus, inspection range of the defect D can be enlarged by writing inspection data in the sate where the head 3 is shifted 8 from the specified position of respective tracks T in the disk at the time of the write step. Thus, the disk storage device itself can rationally inspect the defect without using a private inspection device.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention is a method for inspecting a recording medium of a disk storage device such as a fixed disk device for defects, and the present invention provides a method for inspecting a recording medium of a disk storage device such as a fixed disk device without using a dedicated inspection device. Concerning how to inspect for defects yourself.

[Conventional technology]

As is well known, a thin film of magnetic material such as ferrite, chromium, or cobalt is attached to the disk surface of a disk storage device by plating or vapor deposition as a recording medium, but this recording medium has microscopic particles in the thin film. It is inevitable that some defects due to uniformity or contact scratches with foreign objects will occur, therefore,
In addition to eliminating defective products through precise surface inspection of individual disks, we also inspect the entire surface of the disk for defects after it has been assembled into a disk storage device, and those with a large number of defects are considered defective. The following disks are formatted to exclude defective areas so that they can be used as virtually defect-free disks.

Conventionally, dedicated inspection devices such as defect testers have been used to inspect disks incorporated in devices for defects. This dedicated inspection device first applies a high-frequency write signal to the disk storage device to rewrite the entire surface of the disk, then reads it out and accurately evaluates the analog value of each peak in the signal. This system registers areas where the quality does not reach a specified level as defective areas. Further, after the written content is erased, the remaining content is read out and a location where the analog value of the signal is higher than a specified level is also registered as a defective location.

[Problem to be solved by the invention]

The defect inspection method using such conventional dedicated equipment has the advantage that minute defects can be detected by increasing the frequency of the write signal, and the defect control level can be arbitrarily set by selecting a specified level for the analog value of the successive signal. There is, but
Defect inspection takes time and effort, and the number of disk storage devices that can be inspected with one device is limited, which increases the device cost and makes it unsuitable for mass production.

Another problem with such conventional defect inspection methods is that no matter how high the defect detection accuracy is, the inspection results cannot necessarily be used as a basis for determining whether or not each disk storage device has an acceptable defect. . In other words, whether or not a certain defect is acceptable is determined not only by the degree of the defect, but also by the modulation method used to record data in the disk storage device, the disk formatting method, and the performance of the read signal processing circuit system.

This is because it depends on various factors such as the positioning accuracy of the head operation system.

In addition, with recent technological advances, the number of disk defects, especially the number of major defects, has decreased, and now most major defects are such that acceptance or rejection is determined by the overall performance of the disk storage device. This could lead to being unnecessarily harsh or too lenient.

In this way, for mass-produced disk storage devices,
There is a need for a defect inspection method that better matches the performance of each device and that can simultaneously process a large number of devices without much effort.

In view of the current situation, the present invention provides a practical recording medium defect inspection method that can perform a rational defect inspection that best matches the performance of the disk storage device itself without using a dedicated inspection device, and that is especially suitable for mass production. The purpose is to obtain.

[Means to solve the problem]

This purpose, according to the method of the present invention, includes a writing step of writing test data to each sector with the head placed at a position shifted from the normal position of each track in the disk, and a writing step of writing the test data to each sector, and reading the test data and detecting the read error. A read inspection step that inspects for defects based on the presence or absence of defects, or a write step that places the head on each track in the disk and writes inspection data to each sector, and a read pulse train that reads this inspection data and synchronized with it. Combining this with a reading step of reading inspection data with the timing shifted between clock pulse trains and inspecting for defects based on the presence or absence of reading errors, and registering sectors in which defects are detected in the disk in either case. achieved by.

[Effect]

A feature of the present invention is that the writing step and reading inspection step described above are loaded or incorporated into a disk storage device, and the disk storage device itself automatically reads the recording medium of the disk without using a dedicated inspection device. Furthermore, all defect inspections are carried out sector by sector within each track set during disk formatting.

Therefore, the test data written to each sector in the write step of the method of the present invention does not have special contents as in the conventional method, but is all normal data that is read and written by the disk storage device, and therefore, this data is also used in the read test step. is read as normal data, and the presence or absence of defects within the sector is inspected based on the presence or absence of reading errors at that time. Therefore, in the present invention, defects can be detected using standards that correspond to the performance of the disk storage device, and defects can be detected using practical and rational standards without the risk of being unnecessarily strict or too lenient as in the past. .

However, when defect inspection is performed sector by sector like this,
Conventionally, the disk was not inspected in detail over the entire surface, so defects were mainly detected within the track where the sector was set, and if a defect in the area adjacent to the track was overlooked, the disk When the positioning accuracy of the head decreases due to an unexpected cause while the storage device is in use, a reading error may occur. To solve this problem, it is possible to widen the defect inspection range by writing the inspection data with the head corrected at a position shifted from the normal position of each track on the disk during the writing step as in the above configuration. can.

Furthermore, since the inspection data is read by the read signal processing circuit system of the disk storage device in the read inspection step, even if the waveform of the read signal is slightly distorted due to a minor defect, as long as it can be absorbed within the signal processing circuit system. This defect may be overlooked because the data is read as normal data, and if the operation of the signal processing circuit system differs from that at the time of inspection while the disk storage device is in use, a read error may occur.

To solve this problem, in the read inspection step, as described in the above configuration, the timing between the read pulse train used to read the test data and the clock pulse train synchronized therewith is shifted, so that the test data is read in a state that is intentionally made difficult to read. As a result, even minor defects that slightly disturb the read signal waveform can be detected with high accuracy.

Of course, if you combine the read inspection step, which shifts the timing of this pulse train, with the write step, which writes the inspection data at a position offset from the track described above, virtually all defects, including minor defects, can be detected within almost the entire area of the disk. can be detected with high precision.

〔Example〕

Hereinafter, specific embodiments of the method of the present invention will be described with reference to the drawings. FIG. 1 is a configuration diagram of a disk storage device showing parts related to implementation of the recording medium defect inspection method according to the present invention. In the embodiment described below, it is assumed that the disk storage device is a fixed disk device.

In the figure, a disk storage device 80 is shown surrounded by a dashed line, and normally a plurality of disks 1 are driven by a spindle motor 2, and in this example, a servo motor is installed on each surface of the disk to detect the position of the head 3. Information 51 is written within the fan-shaped area, and several hundred tracks T are set by formatting the disc 1 based on this information.

In this embodiment, it is assumed that a defective sector detected by the defect inspection according to the present invention is registered in the outermost track TO among these.

A head 3 provided for each disk surface is supported by a carriage 4 that is guided through a thin leaf spring so as to be movable in the left-right direction in the figure, and is controlled by a head operating motor 5 that is mechanically tightly coupled to the carriage 4. The radial position on the disk l is manipulated. In this example, a two-phase stepping motor is used as the head operation motor 5, and by applying phase currents Ia and Ib of predetermined polarity and magnitude from a drive circuit 5a for the motor, the head 3 can be moved to any diameter of the disk 1. It shall be possible to position the device in any direction.

Figure 2 shows the key points of this positioning using two phase currents Im and Ib.
In this embodiment, operation command data 00 is given to the drive circuit 5a, and the vector V on the electrical angle created by
By specifying any one of the reference vectors V, the head 3 can be placed on the track T corresponding to the reference vector, and a vector vE or vL that is behind the reference vector V can be specified. Assume that the position of the head 3 can be shifted from the track T by a desired amount.

All the heads 3 are connected to the read/write circuit 6, and the head designated by the head selection command Is is placed in a reading or writing state in response to the read/write command R-. The demodulation circuit 7 shown below includes a peak detection circuit, etc., and the read output R of the read/write circuit 6 described above.
receives an analog read signal R3 from the R3, detects the position of the peak in the waveform, and converts it to a predetermined modulation method 2, for example, M
It is converted into an FM read pulse train signal R5O and output to the encoder/decoder circuit 9.

The off-track detection circuit 8 receives the above-mentioned analog read signal R3 and an index pulse IX generated from a pulse generator 2a built in the spindle motor 2 in synchronization with the rotation of the disk 1, and detects the read signal FIS. An off-track amount S representing the deviation of the head 3 from the normal position on the track T is detected from successive portions of the servo information SI included in the index pulse IX.

The encoder/decoder circuit 9 receives the read pulse train signal 11S.
For example, in cooperation with the synchronization separation circuit IO,
At the same time, the data pulse train signal NS is decoded into a data pulse train signal N5 of a predetermined modulation method, and conversely encoded into a write signal -5 of a predetermined modulation method, which is applied to the write input W of the read/write circuit 6.

The synchronization separation circuit 10 receives the read pulse train signal R3O and separates therefrom the clock pulse CP and another read pulse train signal 1? by PLL operation as usual. sl is output to an encoder/decoder circuit, and in this embodiment, a pulse train shift circuit 70, which will be described later, is incorporated in this for implementing the present invention.

A processor 20 is incorporated in the disk storage device 80 for its internal general control as usual, and in connection with the implementation of the present invention, the drive circuit 5a receives operation command data DO, and the read/write circuit 6 is the head selection command H5
and a read/write command 1111, and a timing shift command TS to the pulse train shift circuit 70 in the synchronization separation circuit 10.
and slow speed command EL, respectively, and the off-track detection circuit 8 reads the off-track amount S in synchronization with the index pulse IX.

The processor 20 also has an internal bus 21. It is connected to a host computer (not shown) via an interface circuit 22 and an external bus 23.

The data control circuit 30 is also a simple special processor that converts the serial data pulse train signal NS received from the encoder/decoder circuit 9 into parallel data and stores it in the built-in RAM 3.
1, or conversely, converts the data into a data pulse train signal MS and supplies it to the encoder/decoder circuit 9, and is connected to the processor 20 via the communication bus 24.

In addition, this data control circuit 30 is connected to its RA as usual.
For example, track 1 read from disk 1 in M31
After temporarily storing data in each sector and verifying whether or not the data in each sector has been read correctly using an error check code in each data, the processor 20 checks whether there is a read error or not via the communication bus 24. Note that the RAM 31 is connected to the internal bus 21 for receiving and passing write data and read data to and from the computer.

The software for the writing step 40 and the reading verification step 60 in the method of the invention is loaded into the processor 20 as shown and performs a series of operations that will be explained later with reference to FIG. Reading inspection step 6
The software for 0 is designed to operate in cooperation with the aforementioned pulse train shift circuit 70 in the synchronous separation circuit 10. Further, in this embodiment, the writing step 40 performs the aforementioned operation of shifting the head from the trunk, and the reading test step 60 performs the aforementioned operation of shifting the timing of the pulse train.

3 and 4 illustrate the meaning of the operation of shifting the head 3 from its normal position on the track T in the writing step 40 of the present invention. FIG. 3 shows an example of the position of a defect or scratch on the track T, and it is assumed that the defect happens to exist from the track T to its adjacent portion as shown in the figure. FIG. 4(a) shows the waveform of the above-mentioned analog read signal R5 read when certain inspection data is written on the track T, and shows the constriction of the waveform corresponding to the defect in FIG. We will now know that this is occurring.

FIG. 4(b) shows the corresponding read pulse train signal R5O from the demodulation circuit 7 of FIG. is detected and a pulse is generated as shown in the figure (in the example of grapes, the waveform disturbance of the read signal F!S due to a defect is near its zero cross point, so the read pulse train signal R3O in the figure (b) Each pulse within is generated at the correct timing without being disturbed by such waveform disturbances.

FIG. 4(C) shows the read signal RS when inspection data different from that in FIG. It's happening nearby. Therefore, the read pulse train signal 1 in the same figure (a)? There is a risk that the timing of the SO pulse may be disrupted, but in this example, as shown in Figure 3, only a portion of the defect is on the track T, so there is relatively little disturbance in the waveform of the read signal R3, and the read pulse train Each pulse of signal RSO occurs at the correct timing. In this way, a defect is easily overlooked when only a portion thereof covers the track T.

FIG. 4(e) shows the read signal R5 when the same inspection data as in FIG. 4(C) is written at a position T3 shifted by δ from the track T as shown in FIG. 3 and read. Since the position Ts almost coincides with the position of the defect, the waveform of the read signal IIS is greatly disturbed, and the read pulse train signal R5O in FIG. A false pulse R5e with a shifted timing is included.

Such a false pulse R3e in the read pulse train signal R5O is
If the timing deviation exceeds the limit that can be compensated within the encoder/decoder circuit 9, a reading error will occur, and this fact will be detected by the data control circuit 3o. In this way, in the method of the present invention, the writing step 4
By causing the head to move from its normal position on the track, defects in areas adjacent to the track can be detected on a practical basis consistent with the data reading performance of the disk storage device.

FIG. 5 shows a specific example of the configuration of the synchronization separation circuit 10 shown in FIG. A clock pulse CPO is established which is synchronized with the read pulse of the synchronized data in the data contained therein as usual. In an ordinary synchronization separation circuit, the read pulse train signal R3O is synchronized with the clock pulse CPO by the synchronization circuit 12, and the read pulse train signal R5I is synchronized with the clock pulse CPO.
However, in the present invention, a pulse train shift circuit 70 is incorporated to mutually shift the timings of both pulse trains.

A delay circuit 71 in the pulse train shift circuit 70 receives the read pulse train signal R5O and delays it by a delay time d, for example, 10μ.
A read pulse train signal R3d delayed by s is outputted and given to the synchronization circuit 12, and another delay circuit 72 receives the clock pulse CPO and generates two clock pulses delayed by delay times d and 2d.

The AND gate 73 is enabled via the attached inverter 73a when the timing shift command TS is 0,
A clock pulse CP is output by delaying the clock pulse CPO by a short time d. The AND gate 74 has a timing shift command TS of 1 and a slow speed command [! When L is O, it is enabled via the inverter 74a and outputs an index pulse CP that is delayed from the clock pulse CPO by a delay time of 2d. Similarly, the AND gate 75 is enabled when the timing shift command TS is 1 and the slow speed command BL is 1, and outputs the clock pulse CPO as it is as the clock pulse CP. Clock pulses CP output from these three AND gates 73 to 75 are output from the synchronous separation circuit 10 via an OR gate 76.

The synchronization circuit 12 receives the above-mentioned delayed read pulse train signal I.
It creates a read pulse train signal R5I synchronized with the clock pulse CP to be given from the IS to the encoder/decoder circuit 9. For example, it is a D-type signal that is set at the rising edge of the read pulse train signal R5d on the condition that the clock pulse CP is a signal. Consists of a flip-flop and associated circuits.

FIG. 6 explains the meaning of shifting the timing between the read pulse train and the clock pulse train by the pulse train shift circuit 70 as described above in the reading inspection step of the present invention. Figure (a) shows the original read pulse train signal R.
3O, and assume that the timing of generation of the third pulse in the figure is shifted from the original position indicated by the broken line due to a defect.

FIG. 4B shows the read pulse train signal R3d delayed by the delay circuit 71 by a time d. The figure (C) shows the clock pulse CP when the timing shift command TS is 0, and it is approximately in the middle of the period when the timing shift command TS is h.
Regular pulses of the read pulse train signal R3d (b) are generated.

FIG. 6 (OFF) is the read pulse train signal R5l that is output when the timing shift command TS is 0.

The rising edge of the pulse that is shifted due to a defect in the read pulse train signal R5d is also the h of the clock pulse CP in the same figure (C).
Since it is within the period of , the read pulse train signal R3d becomes the read pulse train signal R3I as it is.

FIGS. 6(e) and (f) show timing shift command R.
Clock pulse CP when 3 is 1 and slow speed command EL is 1
The clock pulse CP in this case is ahead by a time d than in the case of FIG.
d is output as is as the read pulse train signal R5I.

Figure 6 (chain and (b)) shows the timing shift command R3.
shows the clock pulse CP and read pulse train signal R5I when 1 is 1 and the slow speed command EL is 0. In this case, the clock pulse CP is longer than the time d in the case of the same figure (C).
As can be seen from the figure, the shifted pulse in the read pulse train signal R5d rises before it reaches h.

Therefore, in the read pulse train signal R5I in this case, pulses corresponding to the shifted pulses in the read pulse train signal R5d are missing, as shown by broken lines in the figure. This missing pulse R5
The read pulse train signal R3I with f is decoded by the encoder/decoder circuit 9 and then sent to the data control circuit 30.
is given, but of course a reading error is detected thereby.

As will be seen from now on, by shifting the timing between the read pulse train signal R3d and the clock pulse CP back and forth by an appropriate time d, the original read pulse train signal R5 is restored.
A minute defect can be detected from a slight timing shift of the pulse during O with a desired accuracy depending on the setting of the time d.

Now that we have finished explaining the configuration example of the disk storage device related to the present invention and the role of its writing step and reading inspection step, we will next explain an example of its overall operation with reference to FIG. In FIG. 0, operation steps as a write step 40 and a read inspection step 60 are each shown surrounded by a dashed line.

In the first step S1, flag F, which controls the flow of operations, is set to 1 to indicate that an inspection operation is in progress, and then in step S2, formatting is performed on the entire surface of the disk.
In addition, O is entered in the variable S for specifying the off-track amount of the head.

In addition, during the above formatting, the head is sequentially placed on each track on the disk starting from the outermost track TO, which is normally track 0, and the synchronization separation circuit IO is applied to each track in its entirety. Typically several dozen sectors are configured by writing formanting data, including the synchronization data to be used. Although defects may exist in the writing range of such formatting data, in this example, the writing step 40 includes a step of detecting the defects.

In the next step S3, it is determined whether the flag F is 0 or not, but since it is currently negative, the operation moves to step S4, and the change/failure n, which represents the number of defective sectors in the disk detected by the method of the present invention, is reset to 0. In addition, 0 is entered in the variable l representing the number of inspections.

In this embodiment, the inspection is repeated while sequentially editing the inspection data, so the above-mentioned variable 1 indicates the number of times. The value of this variable l is incremented by 1 in the next step S5, thereby designating the first test. In the following step S6, the flow is switched according to the value of this variable 1, but since the value is 1 now, the operation is at step S7.
Moving on to the data DB with the highest write signal frequency as the inspection data TD, for example, if the data recording is in the MFM1 mode, 1
Specifies repetition of 00 or FF in hexadecimal notation.

The flow then enters operation as a write step 40,
In the first step S40, O is entered in the variable j representing the number of the track on which the head should be placed, and the next step 34 is 0 to specify that the operation should start from the outermost track.
In 1, the head is sought or moved onto the track Tj specified by the variable j, and then shifted from the normal position of the track by the amount specified by the above-mentioned variable S.
Currently, the specified value of this variable S is O, so the head is positioned at the normal position on the track.

In the next step 342, the read/write circuit 6 is placed in a reading state by the read/write command R11, and the formatting data FD written in the previous step S2 is read into the data control circuit 30. The data control circuit 30 stores this data for one track in its RAM 31, verifies whether or not it has been read correctly, and notifies the processor 20 of the presence or absence of a read error and the defective sector number if there is an error. In the following step S43, the operation is switched depending on the presence or absence of this error.

If there is an error, the operation moves to step S44 and it is determined whether or not the flag F is O. However, since flag F was set to 1 in the first step S1, the operation moves to step S45 to determine if there is a defect. The detected sector number is stored in the processor 20, and 1 is added to the change n representing the number of defective sectors.
If there is no error, the operation moves to step 346 where it is similarly determined whether the flag F is 0 or not, but the judgment result is of course negative, and the operation is continued. moves to step 5-47.

The operations of steps S42 to S46 described above are for inspecting defects within the range in which formatting data is written by utilizing the flow of operations of write step 4o as described above.

Step S47 is the original operation step as writing step 4o, in which the test data TD specified in step S7 is written to each sector in the track via the head positioned on the desired track in step S41. , At this time, the test data TD is sent to the data control circuit 30.
The test data TDO write signal -5 is first applied to the read/write circuit 6 from the data control circuit 30 via the encoder/decoder circuit 9 while the read/write circuit 6 is in a write state. Further, it is preferable to write this inspection data ↑D only to sectors that were found to be defect-free in the inspections at steps S42 and 343.

In the next step 34B, it is determined whether the track number variable j is smaller than its maximum value J+w. If so, the variable j is incremented in step 349, and the flow of operation is returned to step 341.

As a result, the operations of steps 541-349 are repeated while sequentially moving the head to the next track, and when the test data TD is written to the entire surface of the disk and the value of variable j reaches the maximum value "-" , the operation flow is step 348
The process exits the loop and moves to step S8. In this step, it is determined whether the flag F is O or not, but the determination result is of course negative, and the flow moves to the reading inspection step 60.

In its first step 560, the track number variable j
is returned to 0. In the 0th step S61, the head is sought to the third track Tj specified by the variable j and positioned by shifting the head from its normal position by the value of the variable S. In the next step 362, the head is shifted from its normal position by the value of the variable S. Timing shift command TS
and put 0 in the slow speed command EL. Steps S63 and 364
is the original operation step as the reading inspection step 60, and in step S63, the inspection data is stored in the data control circuit 30 with the pulse train shift circuit 70 in the synchronous separation circuit IO placed as specified by both commands TS and EL. T
D is read and the presence or absence of a reading error is verified.

In step 364, the flow is divided depending on the presence or absence of this error, and if there is no error, the operation moves to step S66, but if there is an error, the defective sector is stored in the processor 2o using its track number and sector number, for example, in step S65. , if it is a newly detected defective sector, 1 is added to the variable n representing the number of defects.

Steps 366 to 369 are steps for switching the shift amount of the clock pulse CP with respect to the read pulse train signal R5I. Initially, the timing shift command TS is O, so the operation moves from step 366 to step S67, and by setting the timing shift command TS to 1 and returning to step S63, the slow speed command HL is set to O9, that is, the clock pulse CP. The operation of reading inspection data and detecting defective sectors can be repeated in a state in which
The operation moves to step S63, and by setting the slow speed command EL to 1 and returning the flow to step S63, the clock pulse C is
The same operation can be repeated with P delayed.

As described above, after repeating the operation of reading the inspection data and detecting a defective sector in the three cases of the normal state, the early state, and the delayed state of the clock pulse CP, the flow loops from step 36B. As long as the track number variable j is smaller than the maximum value j-, the variable j is incremented in step S71, and then the flow returns to step S61, and the operation as the reading inspection step 60 is sequentially performed. Repeat for next track.

When the operation of the read inspection step 60 is completed for the entire surface of the disk and the variable j reaches the maximum value j-, the operation exits the loop from step S70 and returns to the operation shown on the right side of the figure in the write step 40. to go into.

Steps S50 to S53 in the figure are steps for setting the amount of head deviation when writing inspection data, and in the first step S50 it is determined whether the deviation specification variable S is O, but in step S3 this is set to O. Since it remains the same, the operation moves to step S52, where δ is entered in the deviation specification change S, and the flow returns to the first step 340 of the write step 40.

Thereafter, in step S41 of the write step 40, the head is shifted from the center of the track by the value specified by the variable S, and then test data is written, and while it is read in the read test step 60, defective sectors are detected. In step S41 above, the advanced vector vE and the delayed vector v in FIG. 2 are set so that the deviation from the normal position of the track T of the head 3 in FIG.
Operation command data OD specifying L is sent to the head operation motor drive circuit 5a, and the position of the head 3 is controlled in a closed loop so that the off-track amount S read from the off-track detection circuit 10 becomes equal to the value of the variable S.

After completing the inspection in case the head misalignment is positive δ,
The operation proceeds from step S50 to step S52, and since the variable S is positive, the operation further moves to step 353, where the same operation is repeated with a negative deviation -δ specified for the variable S. After the completion, the operation moves from step S52 to step S9, thereby completing the first defect inspection using the high frequency inspection data DH as the inspection data TD.

In step S9, it is checked whether the variable n representing the number of defective sectors is smaller than the permissible limit N, and if not, the disk being inspected is defective, as shown by the broken line in the figure, by means of display or other means. After that, the flow is immediately terminated, but if so, it is further checked in step 310 whether the variable 1 indicating the number of inspections has reached the maximum value 5-, and as long as it is less than the maximum value, the flow is returned to step S5. In this step S5, the variable i becomes 2, so the current operation is from step S6 to step Sl.
1, the process moves to step S12, and low frequency test data, such as repeating data of 00 or 88, is designated as the test data TD.

After the defect inspection using the new inspection data OL is completed, the flow returns from step S10 to step S5 in the same manner as above, and the operation thereafter moves to step S13 via step S6 and step Sll, and thereafter the process is performed randomly. The generated inspection data Dll is specified as inspection data, and defect inspection using this data is performed until the value of the variable l is the maximum value l.
(1) For example, the number is repeated until it reaches 5.

This predetermined number of defect inspections is performed until the number of defective sectors n is the maximum value N.
After the operation is completed without reaching step SIO
Then, the process moves to step S14, and after setting the flag F to O, the flow returns to step S2 to re-format the disk. When the flag F is 0, the second formatting in step S2 is the same as the first one. In contrast, the inspection is performed by excluding defective sectors that have been detected in previous inspections, allowing the disk to be used as a disk with no defects at all from the outside.

In step S3 after this re-formatting,
Since the flag F is 0, the judgment is correct and the operation is at step S.
15, and by then step 346 and step S6
5, the defective sector stored in the processor 20 is transferred to the aforementioned outermost track T of the disk.

to be recorded. This completes all the operations of the method of the present invention, but in this example, in order to confirm the re-formatting using the write step 40, the track number variable j is set to 0 in the next step 316. The flow is entered in step S41.

In this step S41, the variable S is 0 in step S2.
Since the head is set at the normal position on the track Tj, the formatting data FD is read in step S42, and the presence or absence of an error is determined in step S44. Of course, since there is no error, the operation moves to step S47, and further steps S49 and S
Step S4 of the flow while incrementing the variable j through 50
However, if an error is detected, the operation moves to step S45, where necessary abnormality display etc. are performed as shown by the broken line in the figure, and then the operation is terminated.

When the re-formatting has been confirmed for the entire surface of the disk, the flow moves from step 348 to step s8, and since the flag F is 0, the determination result of this step is output as expected, and the entire operation is now completed.

Although the method of the present invention described above takes longer to operate than when using conventional dedicated equipment, since defect inspection can be performed on the disk storage device itself, inspection can be carried out in parallel on any number of disk storage devices, making it easier to mass-produce. You can get a lot of convenience.

The present invention is not limited to the embodiments described above, but can be implemented in various embodiments. For example, in the embodiments described above, inspection data is written with the head placed at a position deviated from the normal position of the track. The read inspection step is used in conjunction with the read inspection step, which reads inspection data with the timing shifted between the read pulse train and the clock pulse train and inspects for defects based on the presence or absence of read errors.
Even if one of the two steps has a simple configuration in which the head position and the timing between both pulse trains are not shifted, the method of the present invention can be implemented without reducing its main effects.

Furthermore, in the embodiment, the position of the head is shifted from the normal position of the track in accordance with the value of the variable S during the reading inspection step. In some cases, it may be more effective to perform a reading inspection with the written inspection data intentionally made difficult to read, in order to improve defect detection sensitivity, as long as the amount of deviation is not particularly large.

Although the embodiment does not have the function of erasing the recorded content using a conventional dedicated inspection device, for example, instead of returning the flow from step SIO in FIG. 7 to step S5, the entire surface of the disk is erased and then step By returning to S2, the method of the present invention can have a function substantially equivalent to the conventional erase inspection.

〔Effect of the invention〕

As described above, according to the method of the present invention, there is a writing step in which inspection data is written with the head shifted from the normal position of each trunk, and a reading inspection step in which defects are detected from the presence or absence of reading errors by reading the inspection data. Alternatively, the test data can be written by placing the head at a regular position on each track and writing the test data, and by shifting the timing between the read pulse train of the test data and the clock pulse train synchronized with it. By combining the reading inspection step of reading and inspecting for defects based on the presence or absence of reading errors, the following effects can be achieved.

(a) It is possible to automatically inspect the disk recording medium for defects in the disk storage device itself without using a dedicated inspection device. Therefore, during mass production of disk storage devices, defect inspection can be carried out in parallel on any number of disk storage devices without being restricted by the capacity of the dedicated device.

(b) While using the read signal circuit system and head position control system of the disk storage device, recording media can be inspected for defects according to the most reasonable standards that match the performance and accuracy of the disk storage device, and inspection standards can be changed from conventional ones. The irrationality of being unnecessarily harsh or too lenient will disappear.

(C) By writing the inspection data by shifting the head and reading the inspection data while shifting the timing between the read pulse train and the clock pulse train, defects in the range including the adjacent trunk can be efficiently detected. Defect inspection can be completed within a short period of time by simply switching the switch several times.

(d) Since normal data can be written and read sector by sector as test data, the method of the present invention can be easily implemented by loading software onto a disk storage device.

The method of the present invention is not suitable for detecting extremely small defects, such as when using a dedicated defect inspection device, or for obtaining information for improving the process conditions of recording media by understanding the distribution of defects. However, as shown above, it is possible to efficiently carry out the defect inspection process during mass production of disk storage devices.
The present invention also provides a highly practical defect inspection method for recording media that can be inspected according to rational quality control standards.

[Brief explanation of drawings]

All the figures relate to the present invention; FIG. 1 is a configuration circuit diagram illustrating the configuration of the main part of a disk storage device in which the recording medium defect inspection method according to the present invention is implemented, and FIG. 2 shows an example in which the head operating motor is a stamping motor. A vector diagram created by phase current when Figure 5 is a circuit diagram showing a specific example of the configuration of a synchronization separation circuit and a pulse train shift circuit, and Figure 6 is a waveform diagram of a read pulse train signal and a read pulse train signal to shift the timing between a read pulse train and a clock pulse train. 7 is a waveform diagram of a read pulse train signal and a clock pulse to explain the effect, and FIG. 7 is a flowchart showing a specific operation example of the writing step and reading inspection step. : pulse generator, 3: head, 4: carriage, 5: head operation motor, 5a: head operation motor drive circuit, 6: read/write circuit, 7: demodulation circuit, 8: off-track detection circuit,
9: Encoder/decoder circuit, lO two synchronous separation circuit,
11: PLL circuit, 12: synchronous circuit, 20: built-in processor of disk storage device, 21: internal bus, 22: interface circuit, 23: external bus, 24: communication bus, 3
0: Data control circuit, 3i: RAM, 40: Write step, 60: Reading inspection step, 70: Pulse train shift circuit, 71, 72i delay circuit, 73 to 75: AND gate, ? 3a, 74a: Inverter, 76: OR gate, 80: Disk storage device, CP: Clock pulse, C
PO: original clock pulse, D: defect, d: delay time,
DD: Head operation command data, DH: High frequency inspection data, DL: Low frequency inspection data, DR: Randomly generated inspection data, δ: Shift in the writing position of inspection data,! ! L: Slow speed command, F: Frazo, FD: Formatting data, H8: Head selection command, i: Variable indicating the number of inspections, l■: Maximum value of variable l, j Nidrak number variable, j-: Variable j maximum value, Ia, Ib: phase current of the head operation motor, x: index pulse, N: allowable limit of the number of defective sectors, n: variable indicating the number of defective sectors,
NS: Data pulse train signal, R5: Read signal, R5d:
Delayed read pulse train signal, FISe: False read pulse, R5f: Missed read pulse, R5 (Unique read pulse train signal, R5I: Read pulse train signal, Q-: Read/write command, S: Off-track amount or head Displacement specification variables, 81-316: Operation steps, S40-S53:
Operation steps constituting the write step, S60 to S71
: Operation step constituting the reading inspection step, T Nitrack, TD=Test data, Tj: 3rd track, T
S: Timing shift command, D3: Shifted track, TO: Outermost track, V: Phase current vector, vE:
Advanced vector, vL: delayed vector, 11S: write signal. 6

Claims (1)

[Claims] 1) A writing step of writing test data in each sector with the head placed at a position deviated from the normal position of each track on the disk, and reading the test data to determine whether or not there is a reading error. A recording medium defect inspection method for a disk storage device, comprising a reading inspection step for inspecting defects, and registering sectors in which defects are detected in the disk. 2) A writing step in which the head is placed on each track in the disk and test data is written in each sector, and the test data is written with the timing shifted between the read pulse train used to read the test data and the clock pulse train synchronized therewith. A recording medium defect inspection method for a disk storage device, comprising a read inspection step of inspecting for defects based on the presence or absence of read errors, and registering sectors in which defects are detected within the disk surface.