Dynamic magnetic information storage or retrieval – Automatic control of a recorder mechanism – Controlling the head
Reexamination Certificate
1998-06-16
2002-10-08
Hudspeth, David (Department: 2651)
Dynamic magnetic information storage or retrieval
Automatic control of a recorder mechanism
Controlling the head
C360S049000
Reexamination Certificate
active
06462898
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to data storage devices such as a disk drive, and more particularly to error detection and correction of servo system data encoded in position error signals on the disk.
DESCRIPTION OF RELATED ART
A disk drive typically includes one or more disks and a head (for each disk surface) mounted on an actuator which is positionable over a selected one of the tracks on the rotating disk. A voice coil motor (VCM) is commonly used to position the actuator on which the sliders containing the heads are mounted. The slider moves either radially or arcuately toward or away from the spindle center of the rotating disk while trying to find or “seek” to a target track. For any selected track, data is recorded upon the track via a “write path” in the system. Alternatively, data is read from the track via a “read path”.
The positioning of the slider is typically controlled by a closed loop servo system. The servo is operative both in placing the read and/or write transducers (heads) over a target track initially (SEEKING) and in assuring that the appropriate transducer is following the track while the data is being read or written (TRACK FOLLOWING). A portion of each disk track is formatted to include at least one and typically several servo fields to allow the servo system to update its head position several times during each revolution. Each servo field is further formatted to include servo data including position error signals (PESs). The PES fields are conventionally bursts of constant amplitude sinusoidal signals which are of equal length and are spaced at fixed intervals. These fields are used to generate analog signals which vary with the position of the read head and allow the precise positioning of the heads over the track. The PES typically consists of three or four staggered bursts of transitions, for example, an A, B, C and D burst. If the measured amplitude of the bursts indicates that the head is not in position, then a control signal is generated to the VCM to adjust the position. A servo system typically includes a servo processor which controls the VCM and monitors the servo data being read from the disk.
One way of organizing disk storage tracks is to divide a track into data sectors, a data sector being a fixed number of bits recorded along the track. The servo data and PES information are typically located along radially arcing lines, but in the commonly used “zone bit recording” (ZBR) method the number and position of the data sectors varies according to the distance that the track is from the center of the disk. (See for example, Hetzler, et al. U.S. Pat. No. 5,210,660). Usually, there are several servo sectors per track. This results in at least some of the servo sectors being located inside data sectors forming a so-called “split data field.” The servo sectors include the PES and servo data including track identification (TID) and other control information.
Conventionally each track has its unique track identity (TID), i.e. the track address, magnetically recorded in the servo data. During a SEEK, while moving radially inward or outward, the servo system reads the TIDs from the servo sectors on a track. The TIDs tell the servo system where the head is at that moment and allows the system adjust its velocity, etc., to navigate to the target track.
A typical prior art format used on servo sectors comprises the following fields:
∥RdWr|AGC|SID|TID|Hd|Cyl|PES∥.
In this format, the SID (Servo ID) character consists of a timing line or mark and locates the beginning of servo sector position information on the track. The TID (track identity) character is typically expressed as a codeword in a reflected binary (also called a “Gray” code). The virtue of a Gray code is that the TID number varies from track to track by only one digit. This assists robust reading of TIDs even when offtrack as occurs during rapid seeking. HD (head) and CYL (cylinder) fields may provide other information.
In the prior art, there have been suggestions for enriching the PES-stored information. Cunningham, “Quad Burst Servo Needing No Sync ID and Having Added Information”,
IBM Technical Disclosure Billetin
, Vol. 33, No. 3B, pp. 198-200, August 1990, describes a quad burst servo pattern in which approximately 8 bits of information can be encoded in the PES burst by phase manipulation of the fundamental sinusoidal frequency which is used by the servo system. However, the phase method results in a loss of amplitude for normal servo detection. The information can include data unique to each burst identifying it as A, B, C or D, for example by using two bits, as well as some of the lower order bits of the track or sector number.
Similarly, Andresen, U.S. Pat. No. 4,195,320, “Record Track Identification and Following”, issued Mar. 25, 1980, discloses a servo system for a floppy disk which uses track address information digitally encoded in servo bursts using a sliding modulo code and also integrates the signals to obtain an analog measure of the read transducer's position in relation to the track.
While enrichment of PES fields and the like in the servo sector can result in enhanced device access performance, it renders the accurate positioning of the head/transducer over the target track vulnerable where the servo sector information is susceptible to corruption by error or erasure. The corruption may occur randomly or intermittently. The sources may be induced noise, phase jitter, or the like.
SUMMARY OF THE INVENTION
A data storage system with an improved servo system and an improved disk, as well as, a method of recording and retrieving servo information with protection from errors is described herein. The method of the invention involves mapping the predetermined data, e.g. servo data, into codewords with redundancy information such as in an error correction code (ECC) and then recording these codewords into the PES fields. The PES fields will typically be recorded on the disk at the time of manufacture and will not be changed thereafter. The preferred ECC encoding method is selected to maintain analog signal characteristics regardless of the data stored in the PES, so that minimal, if any, change in the conventional servo detection circuitry is needed use the PES bursts as in the prior art, as well as, sources of digital data. Thus, the codewords of the ECC are constrained in the allowed weights, e.g., the number of 1's in each codeword (or magnetic transitions), and run-length-limited (RLL) attributes.
In a preferred embodiment the servo data mapped into the ECC codewords includes track identification (TID) which can be represented by a fixed number of bits of information. The bits are converted into one or more ECC codewords. Additional constraints such as equal weight are placed on the code by mapping the data comprising the TID into limited set of bit patterns (bit-vectors). If disallowed bit patterns are read back, then an error is detected independently from the ECC syndrome calculation process. Placing the TID information in the PES fields using the robust method of the invention optionally allows the conventional TID field in the servo data to be reduced in size or eliminated with a resulting increase in recording capacity. Although various types of ECC algorithms can be used, either a Reed-Solomon type ECC or a Hamming type ECC are preferred.
In one aspect of the invention, the track ID's are written in the PES fields in a selected pattern in which the track ID for each track is recorded in more than one PES field in more than one group of PES fields. One group of PES fields is centered on the selected track while the other group of PES fields is offset so that it partially overlaps the adjacent track, as well as the selected track. When the drive is operating, the previously recorded ECC codewords in one or more of the PES signals is read and the constrained symbols and the ECC codewords are used to ascertain the presence of errors. If errors are found, correction can be performed up t
Blaum Mario
Kabelac William John
Serrano Louis Joseph
Yu Mantle Man-Hon
Altera Law Group LLC
Hudspeth David
International Business Machines - Corporation
Wong K.
LandOfFree
Disk drive with information encoded in the position error... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Disk drive with information encoded in the position error..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Disk drive with information encoded in the position error... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2988918