WO2002035541A2

WO2002035541A2 - Method for detecting transient write errors in a disk drive by detecting thermal interference during write process

Info

Publication number: WO2002035541A2
Application number: PCT/US2001/049711
Authority: WO
Inventors: John R. Stokes; Michael Dickson; Yiping Ma
Original assignee: Iomega Corporation
Priority date: 2000-10-23
Filing date: 2001-10-23
Publication date: 2002-05-02
Also published as: AU2002232712A1; WO2002035541A3

Abstract

In a magnetic disk drive the integrity of data written on a disk is ensured by writing data to a portion of the disk (30) and, simultaneously detecting the presence and/or the frequency of TI (thermal interference) (34) events. If either is detected an error c ondition is generated. The error condition may be reporting the error to the host which decides whether to rewrite the data or immediately rewriting the data without further instructions from the host.

Description

METHOD FOR DETECTING TRANSIENT WRITE ERRORS IN A DISK DRIVE BY DETECTING THERMAL INTERFERENCE DURING WRITE PROCESS

FIELD OF THE INVENTION

This invention relates to magnetic disk drives for computer systems. In particular, this invention relates to methods for detecting errors during write operations in disk drives, especially those caused by transient increases in flying heights by detecting thermal interference events in the read back signal.

BACKGROUND OF THE INVENTION

Maximizing the reliability of the data in disk drives is a key objective of disk drive designers. Unfortunately, that objective often conflicts with the similarly important need for performance. That is, by increasing the reliability performance measures, other performance characteristics such as the data transfer rate, could suffer. For example, when data is written to the disk drive, the success of the write operation, i.e., whether the media accepted the data is unknown. One sure way to guarantee the data after each write operation is by re-reading the recorded data after each write operation. However, a technique that requires all data written to be read as well would severely degrade the performance of the drive. On the other hand, such verification would ensure high reliability of the data. This verification procedure is referred to as a "write with verify." Write with verify takes a long time to perform. It requires that the data be written, then the disk is rotated one full revolution, the data is read, and then verified. U.S. patent 5,588,007, Ma describes a method for detecting transient write errors based on difficulty in reading pre-recorded information on the disk such as servo marks, ID marks and others. It also discusses that in the event of such difficulty, the drive will automatically read the data it just wrote. If errors were encountered during this write process, the drive will either re-write the data or report errors to the host system which will in most cases issue a rewrite operation. However, the pre-recorded information covers only about 20% of the disk space. Therefore, this technique cannot catch all hard errors caused by bad writes. This technique catches about 30% of non-recoverable data errors in one implementation. In disk drives where (G) MR heads are used, an ID-less format further reduces the effectiveness of the transient error detection method described in U.S. Patent

5,588,007.

K.B. Klaassen, J.C.L. van Peppen, "Electronic Abatement of Thermal Interference in head Output Signals", IEEE Transactions on Magnetics, Vol. 33, No.5, September 1997 and U.S. Patent 5,650,887 Dovek, et al. describe what is referred to as (G)MR Technology which is commonly used in heads for hard disk drives. (G)MR heads have been used in rigid disk drives, where a read transducer and an inductive write transducer are built into one slider. (G) MR technology is used to increase the density requirement. This increase in linear density possible with (G)MR Technology makes the "phantom write" problem more severe. A "phantom write is a non-recoverable data error caused by temporary spacing loss during a write. This leads to a greater need for the type of automatic transient error detection method described in U.S. patent 5,588,007.

(G)MR sensors are sensitive to thermal interference (TI) events such as "thermal asperities" and "baseline wander" . "Thermal asperities" result from frictional heating during an asperity-sensor contact, where an asperity may include a disk surface imperfection, external contamination, particles or debris. "Baseline wander" is caused by the cooling of the sensor when the air gap between the sensor and the disk is decreased but not zero. In both cases, large perturbations in the read back signal can be observed. They could lead to non-recoverable data errors and therefore need to be avoided in general. Extensive research and development has been performed by numerous companies and research institutes to understand, eliminate, detect and compensate for the TI events. Pre-amps for the sensors are equipped to detect and monitor TI events during the read process.

It is an object of the present invention to improve transient error detection to make it more effective in higher density disk drives, such as those using (G)MR heads.

SUMMARY OF THE INVENTION

In accordance with the present invention, the integrity of data written on a disk is verified by writing data to a portion of the disk, while simultaneously detecting the presence of thermal interference events using (G)MR sensor, and if, disparities are detected, generating an error condition.

The present invention recognizes the fact that when a particle or contaminant is trapped between the head and media during the write process the resulting temporary spacing loss could lead to non-recoverable data errors (i.e. phantom write). However, when such a particle or contaminant comes into contact with the sensor a "thermal asperity" event occurs and is detectable. In addition, the presence of a third party particle would disrupt the air bearing between the recording head and the disk, causing instability of the air bearing after the exit of the trapped particle. This would lead to the "baseline wander" type of TI event.

According to the current invention, the TI events are monitored during the write process with the existing TI detector in pre-amp chip. A flag is set when a TI event is detected. The data just written could then be immediately read to ensure quality.

In the case where the slider is designed to have constant contact with the media, the sensor is constantly heated by friction. In this case, a "baseline wander" event will be detected when particles enter the head/disk interface causing spacing loss.

In the case where the sensor is designed to make intermittent contact with surface asperities, thermal asperity events could happen at a regular interval. Such a system usually employs two sensors adjacent to each other such that the TI effect can be canceled out during the read process. The frequency of the TI events of an individual sensor can also be monitored during write operation and a flag would be set when a change in this frequency is detected. Such a change in the frequency would indicate spacing loss at the transducer. The foregoing objects, features, and advantages of the invention will be better understood from the following more detailed description and appended claims.

Short Description of the Drawings

Figure 1 is a plan view of a flexible disk, and a flexure with a read/write transducer: Figure 2 shows a read/write head slider for use in a magnetic disk drive;

Figure 3 is a flow diagram of the operation of a disk drive in accordance with the present invention.

Description of the Preferred Embodiment

Figs. 1 and 2 show a magnetic disk drive with a slider mounted on flexures , such as is shown in U.S. Patent, 5,636,085, Jones, et al. A magnetic read/write head assembly includes a slider 11 with transducers 12 and 13 which read and write data recorded on the flexible magnetic medium 14. Write transducer has a recording gap 15 which writes data. The read (verify) sensor 13 is positioned in the direction of movement of media 14 from the recording gap 15. The read sensor 13 is a (G)MR sensor.

In accordance with the invention, transducer 12 is used to write data and transducer 13 is used to detect the occurrence of Thermal Interference (TI) during the write process. Transducer 13 resumes its regular data reading role during read mode.

In general cases, (G)MR sensors are designed to have no contact with recording media surface and/or other particles. Great care is also taken during the design phase of the slider to ensure constant fly height. Since both physical contact and fly height variation could create thermal interference which in read mode could saturate the read back signal making data unreadable, the pre-amp chip of the (G)MR sensor are designed to have internal TI detector such that the data read in the presence of a TI event can be handled accordingly.

According to the present invention, the TI detector is also monitored during the write process. Since the presence of TI event indicates the possibility of fly height variation and/or the presence of external particles at the head/media interface, an error condition will be generated and the write process is suspended as soon as the detection of TI. The write operation will then be repeated when TI is no longer present.

In a preferred implementation, the write operation will continue after the detection of TI. However, a flag will be set. The data just written will be read after the write operation is finished. Any unrecoverable errors could be fixed through rewriting the data at the same location. An additional read operation can be implemented after the rewrite process to ensure the unrecoverable errors are not due to permanent media damage at the error site. In the event of permanent media damage at the error site, the second read operation will still result in unrecoverable error. An auto reallocation operation will then be invoked. During the auto reallocation operation, the bad media site will be flagged and not used in the future. The data will then be written to an alternative site in the disk spare area.

Figure 3 shows a flow chart that describes the process in detail. As indicated at 30, a write request is issued. In response an error flag is initialized as indicated at 32. The TI detector monitoring is activated as indicated at 34. Data is written to the disk at step 36. A determination is made as to whether a TI event occurred during the writing as indicated at 38. If no TI event occurred the process proceeds to branch 40 to determine if writing is complete. If not, the process loops back to step 36 to continue writing data.

If a TI event is detected at 38, a flag is set at 42. The process continues until writing is finished. Then the error flag is checked to see if it is set as indicated at 44. If it is not this is the end of the program as indicated at 46. If the error flag is set, all of the data is read back as indicated at 48. A determination is made as whether the data is valid as indicated at 50. If the data is valid this is the end of the program. If not the data is rewritten as indicated at 52. Again the data is read back as indicated at 54. A determination is made as to whether the data is valid again as indicated at 56. If the data is valid this is the end of the program. If not, the data is reallocated as indicated at 58.

At times, it is necessary to design the recording slider in such a way that the recording element is in constant contact with the recording media. The (G)MR sensor would be subjected to constant heating. In this case only the "baseline wander type of the TI events is possible to occur. Since the (G)MR sensor is subjected to constant heating during the regular operation in this case due to friction between the sensor and the media surface, "baseline wander" type (or the cooling of the sensor) of TI event would indicate the increased spacing between the recording element and the media, which leads to "phantom write error". When "the baseline wander" type of TI event is detected during the write process, the write will either be suspended or be allowed to continue while a error flag is set enabling future read verification at the end of the write operation.

In the case where the sensor is designed to make intermittent contact with surface asperities, thermal asperity events could happen at a regular interval. Such a system usually employs two sensors adj acent to each other such that the TI effect can be canceled out during the read process. The frequency of the TI events of an individual sensor can also be monitored during write operation and a flag would be set when a change in this frequency is detected. Such a change in the frequency would indicate spacing loss at the transducer.

While a particular embodiment has been shown and described various modifications may be made. All modifications within the true spirit and scope of the invention are covered by the appended claims.

Claims

In the Claims:

1. A method of ensuring the integrity of data written on a disk in a disk drive system, comprising the steps of: writing data to a portion of the disk; detecting the presence of TI; generating an error condition if a TI event is detected.

2. The method recited in claim 1 wherein the step of detecting the presence of TI comprises monitoring read back signal.

3. The method recited in claim 1 wherein said step of detecting the presence of TI is performed through a (G)MR transducer.

4. The method recited in claim 1 wherein the step of generating an error condition comprises the further step of re- writing said data to disk.

5. The method recited in claim 1 wherein the step of generating an error condition comprises the further step of reporting an error condition.

6. The method recited in claim 1 wherein the steps of writing and detecting TI are performed by two transducers in close proximity on the same slider.

7. The method recited in claim 6 wherein said transducers are arranged in the direction of the track of data which has been written.

8. The method recited in claim 1 wherein said step of detecting the TI is performed simultaneously with said step of writing.

9. A method of ensuring the integrity of data written on a disk in a disk drive system, comprising the steps of: writing data to a portion of the disk; detecting the frequency of TI events; generating an error condition if a change in the said frequency is detected.

10. The method recited in claim 9 wherein the step of detecting the frequency of TI events comprises monitoring read back signal.

11. The method recited in claim 9 wherein said step of detecting the frequency of TI events is performed through a (G)MR transducer.

12. The method recited in claim 9 wherein the step of generating an error condition comprises the further step of re- writing said data to disk.

13. The method recited in claim 9 wherein the step of generating an error condition comprises the further step of reporting an error condition.

14. The method recited in claim 9 wherein the steps of writing and detecting the frequency of TI events are performed by two transducers in close proximity on the same slider.

15. The method recited in claim 14 wherein said transducers are arranged in the direction of the track of data which has been written.

16. The method recited in claim 9 wherein said step of detecting the frequency of the TI events is performed simultaneously with said step of writing.