KR19990022302A - Apparatus and method for error correction through programmable bias current for magnetoresistive recording head - Google Patents

Abstract

The present invention relates to a data error correction procedure for use in a disk drive comprising a magnetoresistive head. The error correction procedure increases the DC bias current applied to the read stripe of the magnetoresistive head. The increased value is determined for a disk drive with a specific magnetoresistive head. If the conventional error correction procedure is unsuccessful, an increased DC bias current is applied to the read stripe to lower the bit error rate that occurs during the read operation performed by the disk drive.

Description

Apparatus and method for error correction via programmable bias current for magnetoresistive write head

Data read from a disk storage medium is typically read successfully to obtain an accurate representation of the stored data. Occasionally, the bit error rate associated with reading data from a disc can be high due to the increased flying height of the head due to contamination, external stray field and noise, off track noise, etc. Deterioration is caused by an increase in noise that can take shape. Error correction procedures are commonly referred to as play under these conditions. However, error correction involves problems such as data or access time overhead. Thus, the degree of correction is practically limited. Therefore, it is effective to achieve error correction while overcoming or reducing the problem of such error correction.

Summary of the Invention

According to the invention, the apparatus is provided for the detection of errors occurring when reading data or other information from the disc using a magnetoresistive (MR) head. In addition to the MR head, the apparatus includes a read channel that signals the readback signal obtained by the MR head when reading data from the disc. The read channel communicates with a controller portion that performs a number of functions, including properly positioning the MR head relative to the disc to read data from the disc. The controller typically includes a microprocessor or signal processor that processes data.

For using the MR head for its read function, it is well known that a DC bias current must be applied to the magnetoresistive read stripe or element of the MR head. Such bias current is needed to provide a linear response during the readback function. In connection with supplying a DC bias current, in one embodiment, the controller portion communicates with a digital-to-analog converter (DAC) that converts a desired or determined digital signal from the controller portion into an analog current signal applied to the magnetoresistive read stripe. do.

When reading data from the disc using a particular MR head, data errors may occur. If the bits for a particular head deviate from a specified or predetermined bit error rate, one or more error correction procedures are invoked, including, for example, using parity or error correction bits. In particular, in situations where such conventional error correction execution is unsuccessful, excessive time is required and / or multiple error correction attempts are required, so the present invention controllably increases the DC bias current applied to the magnetoresistive read stripe of the MR head. This is useful for error correction. Specifically, as many MR heads indicate, the bit error rate approaches the optimum as the bias current increases. Thus, the DC bias current is controllably increased using a DAC that receives its input from the controller portion.

For the determination of the increased DC bias current, each MR head is checked to obtain a value regarding the increased DC bias current. This value is stored by a disk drive with a particular MR head. The increased DC bias current is useful under certain conditions related to the operation of the disk drive. In particular, during a disk read operation that the user seeks, a bit error rate out of an acceptable range of error rates has occurred, and a decision on whether to call a data error correction procedure should be made. In one embodiment of the invention, in such a determination, one or more standard or conventional error correction procedures are implemented. If conventional error correction execution is not successful, the present invention changes the nominal DC bias current currently being applied to the MR head. Specifically, the previously stored value with respect to the increased DC bias current is used by the controller portion to provide an input to the DAC. The DAC outputs an increased bias current. After the increased bias current is applied to the MR read stripe by the DAC, the disk read operation continues, and the increased DC bias current is part of the error correction procedure in which the error rate is expected to be eliminated or reduced. In one embodiment, at the end of a successful read operation using increased DC bias current, a record is maintained indicating the number of times a particular disk drive has invoked this error correction procedure involving the use of increased DC bias current. . If the number of times that such error correction execution has been used exceeds a threshold, a flag or other indication of the degree of use is provided for access by the user so that this information is provided to the user.

In accordance with the above summary, a number of salient features of the present invention are readily recognized. A data error correction procedure is provided that uses a DC bias current applied to the magnetoresistive head in the disk drive. This error correction technique is particularly useful to supplement other error correction procedures that may not be successful to reduce the bit error rate to an acceptable level. This error correction procedure involving the use of increased DC bias current is by controlling the use of this error correction procedure and in one embodiment by maintaining a usage track to inform the user of data storage problem areas on the disk. Recognize potential adverse effects on magnetoresistive read striations. Preferably, the value of the increased DC bias current used in such error correction procedures is determined during the manufacturing and testing procedures of the disk drive. The value regarding the increased DC bias current for a particular disk drive is stored in disk memory for the next possible use during a disk read operation.

Further effects of the present invention will become more apparent in the following description with reference to the accompanying drawings.

The present invention relates to controlling bias current applied to a magneto-resistive (MR) write head, and more particularly to increasing such bias current as part of an error correction procedure. will be.

1 is a block diagram of the present invention.

2 is a graph demonstrating the effect of increased DC bias current, such as on reducing bit error rate in a disk drive with a magnetoresistive head.

3 is a flow chart illustrating the steps involved in determining an increased DC bias current for an MR head.

4 is a flow diagram illustrating steps relating to an error correction procedure involving the use of increased DC bias current of a particular MR head during a disc read operation.

Referring to FIG. 1, a block diagram illustrating certain basic assemblies or components associated with the description of the present invention is shown. The disc drive 10 includes a disc or magnetic recording medium 14 on which data or other information is stored for a desired retrieval. Data is stored or recorded on the disc using the magnetoresistive (MR) recording head 18. In an embodiment of the invention, the head 18 is a magnetoresistive (MR) read stripe or read element 22 useful for reading stored data or other information from the disk 14 when energized or activated. It includes. In one embodiment, the MR head is connected to an actuator assembly (not shown) of the disk drive 10 that can rotate so that the head 18 is positioned relative to the disk 14.

The output of the MR read stripe 22 is through a read channel 26 that receives a readback signal generated by the MR read stripe 22 during a read operation. Read channel 26 is typically a number of signal processing components, such as preamplifiers, equalizers, filters, and / or peak detectors that process readback signals for subsequent interpretation, processing, or other use by controller portion 30. Or an assembly. For example, the controller unit 30 includes a microprocessor or signal processor useful for the execution of an application program that uses the data read from the disk 14 in the form of a processed readback signal.

The controller section 30 is also substantially involved in the positioning of the MR head 18 relative to the desired track of the disk 14. For example, during a disc read operation, the controller unit 30 accurately positions the head 18 with respect to a desired track by receiving and interpreting servo information from the disc 14 and controlling the position of the actuator assembly.

The controller unit 30 is a disk memory 34 normally used by the controller unit 30 to store data or information that can be easily and quickly accessed by the controller unit 30 in the execution of its desired function. In electrical communication with. The disk memory 34 is usually a semiconductor structure memory, and may be a random access memory (RAM).

In connection with a disk read operation using the MR read stripe 22 to read data on the disk 14, the nominal DC bias current is applied to the read stripe 22 such that a linear response of the readback function is provided when a readback signal is generated. It must be authorized. In one embodiment, the DC bias current is applied to the read stripe 22 using a digital-to-analog converter (DAC) 38 or other current control circuit. The DAC 38 communicates with the controller unit 30 as shown in FIG. The controller unit 30 applies a digital signal representing the DC bias current to the DAC 38, which converts the DC bias current into an analog current signal applied to the read stripe 22, thereby executing the read operation. Activated in connection with the It is well known that the current density in the read stripe 22 is related to its effectiveness or working life. Specifically, a relatively high DC bias current level results in a dangerously high current density in the read stripe 22, which in turn causes the stripe 22 to fail. This failure mode can occur through electron transfer or more generally through thermal failure. On the other hand, a relatively high DC bias current is useful in providing an optimal timing margin or bit error rate, as can be found in many MR write heads. This is shown graphically in FIG. 2, which plots the timing margin or bit shift distribution for a particular MR head when reading a composite pattern of data using the read stripe of the particular head.

As can be seen in FIG. 2, a number of curves are shown, each associated with a different, specific DC bias current applied to the read stripe. Two individual curves or graphs are shown in FIG. 2, graph A represents data with a DC bias current of 12 mA and graph E shows data associated with a DC bias current of 5 mA. As can be appreciated, for a given bit shift, as the bias current increases, the error rate decreases, and the bit shift is a function of the composite data pattern. For example, for a bit shift of 4.1 nSec, the data error rate is 10 ⁻⁶ when the DC vise current is 12 mA and the error rate is about 10 ⁻³ when the DC bias current is 5 mA for the same bit shift. . Thus, with respect to this example, for the same composite data pattern using the same MR head, the error rate is reduced or improved by 1,000 (10 ³ ).

Based on the float of FIG. 2, it is desirable to substantially increase the DC bias current applied to the MR read stripe in order to reduce the data error rate, but this results in undesired high current density in the stripe, thus finding and realizing a tread-off. Should be. This is preferably realized by using the increased DC bias current as part of the error correction procedure in the present invention. In particular, during read operations by disk drives with specific MR heads, in some circumstances the DC bias current applied to the read stripe changes from nominal size to increased size to improve or improve the readback function and optimize or reduce the error rate. do. Preferably, the change to increased DC bias current is made after one or more standard or conventional error correction procedures are invoked and unsuccessful. Also preferably, the increased DC bias current is applied to the read stripe 22 for a relatively short period of time, which period depends on the magnitude of the increased DC bias current. In one embodiment, the increased DC bias current is in the range of 10-20 mA, and the range of time such increased DC bias current is applied is in the range of one sector or complete rotation, for example about 0.1-15. It is in the range of mSec.

Referring to Fig. 3, the determination of the increased DC bias current for a particular MR head will be described. According to block or step 50, the disk drive is tested to determine whether the increased DC bias current is appropriate as part of the error detection procedure and, if appropriate, the value of the increased DC bias current. For a disk drive with a specific magnetoresistive head, while the nominal DC bias current is applied to the read stripe, the head begins to read the complex data pattern stored on the disk as part of the test procedure. During the reading of the composite data pattern, it is determined whether error correction is required in step 54. In this regard, it is determined whether there is an acceptable data error rate during reading of the composite data pattern. In one embodiment, the acceptable error rate is less than 10 bits in error of 10 ⁷ bit reads.

If no error correction is required in the determination in step 54, then in step 58 it is checked whether the composite data pattern has been read out sufficiently or in a predetermined number of disk rotations or times. Such a predetermined number is about 10 ⁶ total bits, which is usually a sufficient number leading to the error correction required for most magnetoresistive heads. However, if the composite data pattern has been read a sufficient number of times and the error rate does not exceed an acceptable level, then determine that no increase above the nominal DC bias current is necessary for this particular magnetoresistive head and test the next disk drive with another magnetoresistive head. do.

In most cases, however, the composite data pattern being read produces an exceeding acceptable error rate for a particular magnetoresistive head, and at step 62 the current DC bias current is increased by a certain amount. Initially, this increment increases the DC bias current from its nominal magnitude. The predetermined magnitude is the amount that will effectively affect the applied DC bias current so that there is a significant expectation that the complex data pattern can be read without requiring error correction. Thus, in step 66, the composite data pattern is read back from the disk using the previously increased DC bias current. That is, in the embodiment of FIG. 1, the output of the DAC 38 is applied to the read stripe as a current that has been increased by a predetermined magnitude above the nominal DC bias current.

During the reread of the composite data pattern, according to step 70, it is determined whether there is now an acceptable error rate for reading data from the disc. If the error rate is determined to be unacceptable and error correction is required, then at block 62 the current DC bias current is increased again and at block 64 the composite data pattern is read back again. On the other hand, if error correction is not required, in step 74, it is checked whether the composite data pattern has been reread for a predetermined or sufficient number of times for the purpose of generating an unacceptable error rate during the read operation. If such a predetermined number of times is not exceeded, according to step 66 the composite data pattern is continuously read back from the disc. However, if the composite data pattern has been read for a sufficient number of times, step 78 is executed. That is, the current magnitude of the DC bias current is increased by an adjustment factor such that an increased DC bias current is obtained. The adjustment factor increases the previously determined DC bias current by a certain amount or percentage. In one embodiment, the adjustment factor increases the determined DC bias current by about 10 percent. This percentage increase is such that when the increased DC bias current is applied to the read stripe during a read operation that called the increased DC bias current as part of the error correction procedure, the previously increased DC bias current effectively meets the desired or acceptable error rate. Provide a safe coefficient or margin to ensure that. When the final increased DC bias current is reached according to the combination of the adjustment factor and the previously determined DC bias current, in step 82, it is stored in the disk memory for potential access by the disk controller during the error correction procedure.

Of course, other methods or steps may be used to determine the increased DC bias current. In another embodiment, instead of merely increasing the current DC bias current by a predetermined magnitude, one can reduce the current DC bias current. In such a case, it may be useful to increase the current DC bias current by a relatively large amount, and if such a current DC bias current produces an acceptable error rate, the DC bias current should be reduced. This reduced DC bias current can be applied to the read stripe to determine whether a particular DC bias current produces an acceptable error rate. According to such a decision, even if it is not larger than necessary, the current DC bias current can be further increased or decreased until the current DC bias current achieves an acceptable error rate. When reading a composite data pattern for a predetermined number of times or during disc rotation, such a pattern can be stored in succession in a disc track so that the head does not inherently move when the disc moves relative to the head. In addition, the composite data pattern may be stored such that the movement of the head relative to the disc during the reading of the composite data pattern, and the movement of the head off the track during the adaptation of the bias current, compress the system SNR.

When the increased DC bias current is stored in the disk memory for a disk drive having a particular magnetoresistive head, the steps associated with the use of the increased DC bias current as part of the disk read operation are described with reference to FIG. According to block or step 86, data or other information is read from the disc using a previously tested magnetoresistive head. During this read operation, a nominal DC bias current is applied to the read stripe 22. In step 90, it is checked whether the error correction procedure is called because of an unacceptable error that occurred during the read operation. If no such error correction procedure is required, it is determined in step 94 whether the particular read operation has ended. If so, an end-of-read operation occurs at block 98. However, if the read operation has not ended, data continues to be read from the disc in accordance with step 86 until the read operation ends or until there is an unacceptable error rate and error correction is required. When error correction is required, in one embodiment, a standard or conventional error correction procedure is called in step 102, and it is checked whether one or more of these error correction procedures were successful in achieving the desired error rate. If so, again in step 94, it is determined whether the read operation has ended, and if not, reading from the disc continues until the end of the specific read operation. On the other hand, if the standard or conventional error correction procedure is not successful, at step 106, the nominal DC bias current applied to read stripe 22 is changed to increased DC bias current. This is typically accomplished by the controller unit 30 accessing the disk memory 34 to retrieve this value and apply it to the DAC 38. The DAC 38 outputs a value relating to the increased DC bias current, which is applied to the read stripe 22 to attempt to correct data errors resulting from a particular read operation in the read stripe 22. Generates. In this regard, at block 110, the read operation continues at block 114 to its end. At the end of the read operation, it is determined at step 118 whether the use of the increased DC bias current has exceeded a predetermined number of times for the particular MR head. The purpose of this decision is to provide the user of this particular disk drive with information about the number of times that increased DC bias current has been called because such increased DC bias current has affected the working life of the read stripe, and thus the proper functioning of the disk drive. To provide. In one embodiment, in step 122, a flag is set for providing this information, which is accessed or provided to the user.

Of course, if the increased DC bias current does not result in an acceptable error rate, the user is informed that the increased DC bias current error correction procedure was unsuccessful and / or the data just read does not meet an acceptable error rate. You can call another error recovery procedure, such as providing an indication.

In other embodiments or variations of the invention, the increased DC bias current is variable. Instead of one stored value for a disk drive with a particular MR head, the DC bias current may increase or change more than once during the error correction procedure. For example, if one increased DC bias current fails to achieve the desired error rate result, another increased DC bias current with a predetermined increment greater than the previously used increased DC bias current is applied. Thus, in addition to or instead of making or assembling a single increased DC bias current value, such an embodiment offers the possibility for one or more such values.

The above description has been submitted for illustrative purposes. In addition, the description is not intended to limit the invention to the form described herein. The foregoing embodiments illustrate the best mode presently known for practicing the present invention and those skilled in the art will use the present invention as claimed or by other embodiments, by various modifications or uses of the invention as required by their particular application. I want to be able to. The scope of the appended claims is to be construed to include another embodiment to the extent conventionally permitted.

Claims (17)

In the method for correcting data read from a disk, Preparing a first magnetoresistive head for reading data from the disc by the first magnetoresistive head having a read stripe; Concluding that error correction should be performed; And Invoking one or more error correction procedures comprising using an increased DC bias current applied to the read stripe Data correction method comprising a. The method of claim 1, The preparing step includes testing the first magnetoresistive head using a data pattern that generates a read error during a disc read operation. The method of claim 1, The preparing step includes determining a value with respect to an increased DC bias current for the first magnetoresistive head. The method of claim 3, wherein The determining step Confirming that data being read from the disc is not being read at an acceptable level; Changing the magnitude of the DC bias current applied to the read stripe of the first magnetoresistive head; Rereading data from the disc using the first magnetoresistive head; Continuing to change the DC bias current applied to the first magnetoresistive head when an acceptable level of error rate is not achieved; And Ceasing said change in said DC bias current when an acceptable error rate is achieved. The method of claim 3, wherein The determining step includes adjusting a previously determined value with respect to the DC bias current using the adjustment factor. The method of claim 3, wherein And the determining comprises storing a value relating to the increased DC bias current using a memory of a disk drive having the first magnetoresistive head. The method of claim 1, The concluding step includes determining that a nominal DC bias current applied to the read stripe should be increased. The method of claim 1, The invoking step starts the first error correction procedure and then when the first error correction procedure fails to reduce the error rate for reading data from the disc using the first magnetoresistive head to an acceptable level, Changing the DC bias current applied to the read stripe to the increased DC bias current. The method of claim 8, And said altering step includes accessing a value relating to said increased DC bias current from disk memory. The method of claim 1, Said invoking includes indicating that said DC bias current has been used in connection with an error correction procedure. The method of claim 1, Said invoking includes continuing to read data from a disc using said first magnetoresistive head after said DC bias current is changed from a nominal DC bias current to said increased DC bias current. An apparatus for error correction during a read operation in a disk drive having a magnetoresistive head, comprising: A disk for storing data; A magnetoresistive head comprising a magnetoresistive read stripe; Controller means in communication with the magnetoresistive stripe to determine that one or more error correction procedures should be invoked; And Increased DC bias current in the magnetoresistive read stripe when communicating with the controller means to determine that the one or more correction procedures should be invoked, wherein the increased DC bias current is such that the one or more error correction procedures will not be invoked. And a control circuit for applying a different value than a nominal DC bias current applied by the control circuit to the magnetoresistive read stripe. The method of claim 12, And said increased DC bias current is predetermined for said first magnetoresistive head. The method of claim 12, The increased DC bias current is an error that is determined using a magnitude that follows reading the data from the disc by the first magnetoresistive head after successfully reading the data from the disc and also using an adjustment factor that increases the magnitude. Correction device. The method of claim 12, And said controller means comprises memory means for storing a value relating to said increased DC bias current. The method of claim 12, And a pattern for storage on disk that requires a DC bias current different from the nominal DC bias current to acceptably read the pattern. The method of claim 12, The one or more error correction procedures include using a first error correction procedure that is different from the increased DC bias current, wherein the first error correction procedure causes the control circuit to apply the increased DC bias current to the read stripe. Error correction device that is executed separately from the prior to authorization.