JPH02103783A - Defect checking device for magnetic disk - Google Patents

Abstract

PURPOSE:To improve the error detection precision by providing a missing signal generating circuit to reduce the level of the output signal of a test data writing circuit or invalidate this output signal at a set timing. CONSTITUTION:A missing signal generating circuit 94 incorporates a preset counter 94a and a driver 94b. The preset counter 94a receives test data sent from a check sequence control circuit 13 to a test data writing circuit 91 and counts down its value, and when the counted value reaches 0, the output signal is generated to the driver 94b. By this output signal, the driver 94b is turned on for a period corresponding to one clock signal and the level of the write data signal outputted to a write/read amplifier 92 is reduced to about zero to invalidate only the write signal of one bit then. Thus, it is easily confirmed whether an error checking circuit system is normally operated or not, and the error detection precision is improved.

Description

Detailed Description of the Invention [Industrial Field of Application] The present invention relates to a magnetic disk defect inspection device (hereinafter referred to as a certifier). The detection status of the defect detection circuit can be checked in the Certifier, which detects missing errors and extra errors (extra errors) "9" as defects by writing the write clock rate as test data or erasing the write test data. Regarding certifiers like this.

[Prior Art] When a magnetic disk device used in a computer/system has an unrelated defect in the magnetic medium, recording of 11) included data becomes incomplete, resulting in error/puffer error or extra error. do.

Eliminate such defects to maintain good quality, and
li+j 1. To ensure this, various inspections are conducted using certifiers during the production process.

One of these is defect inspection of magnetic media.

To inspect magnetic media for defects, first write test data of write clock I'71 (1 bit of continuous bit data, test pattern data, etc.) into the test trunk.
It is read out and judged, and the number and length of defective bits (missing) are determined by 5 Wt to judge whether the track is good or bad. Also, n) delete the input test data and determine whether there is an extra error. Furthermore, the grade is determined based on the number of defective tracks among all the tracks. Then, based on such determination, the magnetic disk is evaluated, and if it has a certain level of defects, it is rejected. In addition, in the case of an initializer that has a certifier inside, it is necessary to allocate an alternative sector or track to an unusable sector or track, or to skip defects, depending on the defective data detected by the certifier. , format the magnetic disk, and initialize the magnetic disk.

By the way, the density of data recorded on a magnetic disk is
-1- The number of defects that the above-mentioned certifier or initializer (including certifier + formatter) must detect and process data has become higher than before. The number of layers is increasing. Furthermore, various lengths of defects have occurred, and it has become impossible to uniformly determine collect errors and uncollect errors. Moreover, the defect detection status is related to the inspection accuracy at that time (
Ru.

[Problem to be solved] Therefore, the degree of inspection accuracy of the certifier is a serious issue, but in order to check the error detection i of the certifier of 17 degrees, it is usually necessary to use a caliper plane. This is done by connecting a phosphor device to the certifier to generate a missing waveform signal, and adding it to the certifier to measure the state of error detection.

However, in such a check method, the caliper line device must be connected to perform the check, and there is a problem in that it takes time and effort to connect and disconnect the caliper line device. Moreover, the calibration device does not generate signals close to the actual detection conditions, such as error signals generated in relation to the magnetic disk, so as the error detection accuracy increases, a discrepancy occurs between these signals. . Further, in a defect inspection with high error detection accuracy, the number of checks of the inspection operation increases.

An object of the present invention is to provide a certifier which can easily test the error detection state of a circuit, and which solves the problems of the prior art.

[Means for Solving the Problems] The configuration of the certifier of the present invention can be used in a magnetic disk defect inspection apparatus that writes test data in a test trunk and reads it out or erases it to detect defects. The data write circuit includes a data write circuit and a missing code generation circuit that lowers the level of the output signal of the write circuit at a set timing or makes it invalid.

[Operation: By providing the certifier with a missing signal generation circuit that disables the output signal of the test data writing circuit at a set timing, missing data can be written at a predetermined position on the magnetic disk under test. By reading this missing data, the certifier itself can perform a missing check at any time. As a result, it becomes easy to check whether the error checking circuit system is operating normally, and there is no need to specially connect a calibration device to the certifier.

[Example Code] Please refer to the drawings for an embodiment of this invention! t(1) and will be explained in detail.

Figure 1 is a block diagram of the embodiment, and Figure 2 is a block diagram of the certifier test data input/readout control circuit according to the present invention. FIG. 3, a block diagram of the /stem, is an explanatory diagram of the missing error interrupt signal pf and read signal. Note that in each of these figures, equivalent components are indicated by the same line numbers, and their explanations are omitted.

In FIG. 2, ■ is a magnetic disk drive device that drives magnetic disks one by one, and a magnetic disk 3 to be inspected is attached to its spindle motor 2, and the magnetic disk 3 is accessed by a magnetic head 4. Writing/reading of data is performed. An encoder 5 is attached to the rotating shaft of the spindle motor 2, and rotational position information is input from the encoder 5 to a rotational position detection circuit 6, and is stored in each track of the magnetic disk 3 by the rotational position detection circuit 6. A clock signal is generated corresponding to one bit of data. This clock signal generated by the rotational position detection circuit 6 is then manually inputted to a ring counter 8 which makes one cycle in 8 bits (octal) of the defect data extraction circuit 7.

The ring counter 8 receives a clock signal generated from the rotational position detection circuit 6 in accordance with the data writing position, counts the clock signal, and completes counting every 8 times.
Output r Shinkyu. Then, this is added to the byte counter 11 as a detection signal for one byte.

The byte counter 11 receives this 1-byte @11 signal, updates its count value, and sends the count value to the defective data memory section 12 as write data.

This byte counter 11 is reset in response to an index detection signal when the rotational position detection circuit 6 detects an index. Therefore, each count value is 0
~n indicates each address obtained by dividing the test track into the 91st byte. When a defect occurs, the address value corresponding to the position is sent to the defect data memory section 12 and stored therein as will be described later.

1! ) Loading/reading control circuit 9 controls a head calibration head (not shown) according to the control of the inspection sequence control circuit 13.
is controlled to position the magnetic head 4 on a predetermined test track, write/read data to that track is controlled, and a read signal is sent to the error detection circuit 10. As shown in FIG. 1, this write/read control circuit 9 receives a test data write circuit 91 and a J) write data signal from the test data write circuit 91 and sends it to the magnetic node 4. A write/read amplifier 92 outputs the i' write signal, and a write/read amplifier 9 reads the signal read from the trunk of the magnetic disk 3 via the magnetic head 4.
Mino/Fugu 4 whose output is connected to the output side of the read circuit 93 and the test data write circuit 91 that receive data via the test data write circuit 91
No. 11 generating circuit 94.

The missing signal generation circuit 94 is activated and operated by a control signal from the inspection sequence control circuit 13, and has a preset counter 94a and a driver 94b therein. Timing data for generating a missing item is preliminarily set in the preset counter 94a from the inspection sequence control circuit 13.
receives the test data sent from the test sequence control circuit 13 to the test data writing circuit 91, counts down its value, and generates an output signal to the driver 94b when the count value reaches zero. With this output signal, the driver 94b connected to the output side of the test data writing circuit 91 is activated for one clock signal period.
ON +1 state is entered, and the level of the write data signal output to the write/read amplifier 92 is lowered to near zero level to prevent it from being output, and only the 1-bit write signal at that time is invalidated.

FIG. 3 shows the invalidated write clock signal and WCK in this case (see (a) in FIG. 3) and the input signal "W1" (see (b) in FIG. 3), The readout signal RD (see (C) in FIG. 3) is also shown.

In this way, by missing the write data, it is possible to generate a min/mining state and write the test booter.
Its position is determined by setting timing data in the preset counter 94a from the inspection sequence control circuit 3, and data having a missing waveform can be written to track 1- of the magnetic disk 3. Furthermore, in this case, it is possible to set the missing waveform at any position on track 1-.

When the error detection circuit 10 detects an error from the readout signal, the error detection circuit 10 outputs an error detection signal St and type data S2 indicating the type of the detected error such as a missing/missing error, a missing error, a modulation error, or an extra error. and sends these signals to the defective data memory section 12.

In addition, in the case of extra error detection, the test data written in the test track is erased and the data is read out, so the error detected at this time is one type of extra error. Become.

A microprocessor (MPU) 15 is controlled by a program and controls the defect data extraction circuit 7 and the inspection sequence control circuit 13 via the bus 14. The inspection sequence control circuit 13 operates in response to a test command specifying a test track scent etc. from the MPU 15, and controls the magnetic disk drive device 1 and the write/read control circuit 9 of the defect data extraction circuit 7.

Therefore, when inspecting defects, first, the magnetic disk drive device 1 and the write/read control circuit 9 are controlled by the inspection sequence control circuit 13 that operates according to the control of the MPU 15, and the i'f clock signal of the 2F frequency is (or other test patterns) is written to a designated test track on the magnetic disk 3 by the magnetic head 4.

-) The loaded test data is read from the test track by the magnetic head 4 and inputted to the error detection circuit 10 via the write/read control circuit 9, where the basic H/level data and the < is a wind pulse (, read as i"y, i1, bow name or its pulsation 4i: Compared to a bow, the above-mentioned minon/"guerer bekunft error,
Various error detections such as a monolayer error and an extra error are performed, and this test 1 error data is stored in the defect data detection memory section 12 together with its position information.

Here, assuming that the MPU 15 collects defect data corresponding to one test track from the defect data memo') 12c, each time data for one round of the test track is stored in the defect data memory section 12, the inspection sequence control is performed. circuit 1
3, the MPU 15 receives the control response signal -l, and in response, the MPU 15 reads defective data for one track from the defective data memory 12 via the bus 14 and transfers the data to the internal memory 16. The data is stored corresponding to the test track.

As a result, at the end of the inspection, the defect data for all tracks on the magnetic disk 3 is stored in a predetermined area of the internal memory 16, corresponding to each test track, and managed at the 11th address. It will be remembered.

Therefore, the MPU 15 reads this defect data stored in the internal memory 16 according to the defect analysis processing program l, and performs defect analysis processing on the defect data managed by addressing this byte i)'f. Can be done.

In this case, since the defective location is the address of pi) jlj, it is necessary to determine whether or not an alternative sector or trunk can be allocated and used for the sector and track, and further detailed bit-by-bit testing is required. It becomes easy to judge whether or not. Even if a track is unusable, it can be easily determined whether it is a usable track or not by checking the bits again. Then, for such a track, detailed defect inspection can be performed only on the sector or track that is to be inspected again. This improves yield by
- The total inspection time and defect data analysis time can be shortened.

When the error detection circuit IO determines whether the IF is always operating by performing such processing with a certifier, the missing signal generation circuit 94 is activated and the missing signal is generated at the position where the missing signal is to be generated. In order for the generation circuit 94 to operate, data is set in the preset counter 94a, and the error detection circuit 10 is set to 1. The magnetic disk 3 may be inspected by setting it to the single error detection state. In this case, as long as the error detection circuit 10 or the defect data extraction circuit 7 as a whole operates normally, a missing error will be detected at the set position, and if not, no missing error will be detected at that position. Note that the driver 94b is capable of operating to lower the level of the output of the test data writing circuit 91 by a predetermined M, and is provided with a register or the like to store the value of this level setting, and is programmed according to the value of this register. The output of the driver can be adjusted.

By installing the miso/blowfish signal generation circuit 94 in this way, even within the certifier, the error detection circuit 1
0 or missing 1 (61 data extraction 111 It is possible to check whether the entire circuit 7 is always operating at 1F.Moreover, it is possible to check in a short time.In particular, the driver 94b can check for missing data. )) By adjusting the output level of the embedded clock signal, it is possible to incorporate 7'J of the missoning waveform into the magnetic disk 3 at different levels of FiTt, and inspection accuracy can be measured at various levels.

As explained above in 1-1, in the embodiment, the type of defect is stored in correspondence with the defect to be detected, but the error detection circuit detects the specified defect in A, like an extra error. In such a case, there is no defect type (i-''') that occurs in the detection circuit. Therefore, the error detection circuit in the embodiment has the function of detecting defects according to the type. good.

In the embodiment, the test track is managed and addressed in 1 byte Il1 position as the position data of the defective trolley 1, and is designated as defect I1 or 6. However, this invention does not necessarily identify the defect (i"! It is not limited to those that are managed.

By the way, the present invention is of course applicable not only to hard disks but also to so-called flexible disks.

[Effects of the Invention] As described in 1- below, in this invention, the missing signal that invalidates the output dip of the test data writing circuit at a set timing can be prevented by providing a generating circuit in the certifier. Inspection magnetic disk 1. It is now possible to insert missing data into a predetermined position of -), and by reading this missing data, the certifier itself can perform a minion/g test at any time. As a result, it becomes easy to check whether the error check circuit system is always operating or not, and there is no need to specially connect a caliplane unit to the certifier.

[Brief explanation of drawings]

Figure 1 is a block diagram of the embodiment, and Figure 2 is an overall system centered on the defect data extraction circuit. Block diagram, 3rd
The figure shows the write signal and readout 43 of the error.
FIG. ■...Magnetic disk drive device, 2...Spindle motor, 3...Magnetic disk, 4...Magnetic head, 5
...Encoder, 6...Rotational position detection circuit, 7...
・Defect data extraction circuit, 8...Ring counter, 9・
...Write/read 11'' control circuit, 10...Error detection circuit, 11...Byte counter, 12...Defect data memory section, 13...Inspection sequence control circuit, 14...Bus, 15...Microprocessor (MP
U), 1B...Internal memory. 9th...Test data writing circuit, 92...Writing/reading amplifier, 93...Reading circuit, 94...Missing signal generation circuit 94°

Claims (1)

[Claims] (1) In a magnetic disk defect inspection device that writes test data on a test track, reads it, or erases and reads it to detect defects, the test data writing circuit and the output signal of this writing circuit are set. 1. A magnetic disk defect inspection device comprising: a missing signal generating circuit that lowers or disables a level at a timing.