Dynamic magnetic information storage or retrieval – Automatic control of a recorder mechanism – Controlling the head
Reexamination Certificate
2001-06-28
2004-07-27
Hudspeth, David (Department: 2651)
Dynamic magnetic information storage or retrieval
Automatic control of a recorder mechanism
Controlling the head
C360S077020
Reexamination Certificate
active
06768609
ABSTRACT:
FIELD OF THE INVENTION
The claimed invention relates generally to the field of disc drive data handling systems, and more particularly, but not by way of limitation, to a method and apparatus for improving servo control response in a disc drive servo system through the iterative calibration of compensation values stored in a compensation table and used to reduce nonlinearities in a position error signal.
BACKGROUND
Disc drives are data handling systems used to magnetically store and retrieve digital data files. A typical disc drive comprises one or more rigid recording storage discs arranged about a spindle motor for rotation at a constant high speed. A corresponding array of read/write heads are provided to transfer data between tracks defined on the disc surfaces and a host device (such as a computer) in which the disc drive is mounted. The heads are mounted to a rotary actuator and are controllably positioned adjacent the tracks through the application of current to an actuator motor (such as a voice coil motor, VCM).
Present generation disc drives typically employ heads which utilize separate read and write elements. The write element typically has a thin-film inductive coil construction with a write gap placed in close proximity to the recording medium. Input data to be written to a disc are encoded and serialized to generate a series of bi-directional write currents which are applied to the write element. Each change in the polarity of the write current results in a magnetic flux reversal, or flux transition, in the recording medium. Data are thus recorded along each track at a selected frequency in relation to the presence of a flux transition (a logical 1) or the absence of a flux transition (a logical 0) at regular intervals along the track.
The read element is typically provided from a magneto-resistive (MR) material which is configured to exhibit changed electrical characteristics when subjected to a magnetic field of a selected orientation. During a read operation, the read element is biased using a relatively small bias current (or bias voltage), and the selective magnetization of the disc is detected in relation to induced changes in voltage across (or current through) the read element caused by the magnetization pattern along the track.
A closed loop, digital servo control system (servo circuit) is used to control the position of the heads. Servo data written to the discs during disc drive manufacturing are transduced by a selected head and provided to the servo circuit to sense head position. The servo circuit generates a position error signal (PES) indicative of head position error and applies current control signals to the actuator motor to adjust the position of the selected head to a desired relation to the selected track (such as over the center of the track). A typical PES is expressed as a sequence of digital PES samples which ideally increase at a linear rate across the width of the track as the selected head is swept from one track boundary to the next.
However, various factors have been found to introduce nonlinearities in a PES so that the PES increases in a nonlinear fashion as the head is swept across the width of a track at a selected rate. Such nonlinearities, if sufficiently pronounced, can adversely affect the stability of the servo loop. Efforts have therefore been employed in the prior art to linearize a PES.
One approach exemplified by U.S. Pat. No. 6,091,567 issued to Hampshire initially determines different forms of PES from the same transduced servo position data, with each of the different, initial PES forms having different nonlinear characteristics across the radial width of a selected track. A final, composite PES is formed as a weighted combination of the various initial PES forms, with the final, composite PES having substantially linear characteristics across the radial width of the track.
Another approach exemplified by U.S. Pat. No. 5,774,297 issued to Hampshire et al. also determines different forms of PES having different nonlinear characteristics form the same transduced servo data. However, this reference precisely maintains the position of the head in one location over the track (such as over track center) using one of the PES forms that exhibits nominally linear characteristics in this location while adding compensation terms to a second one of the PES forms until the second PES form exhibits linear characteristics in said location. The compensation terms are thereafter used in conjunction with the second PES form to position the head during normal operation.
U.S. Pat. No. 6,046,879 issued to Cooper et al. takes a different approach: instead of attempting to adjust a raw PES to remove nonlinearities therein, this reference provides a nonlinear controller which models, and thereby compensates, the nonlinearities present in the raw PES.
It should be noted at this point that the PES nonlinearity problems discussed above are substantially different from, and are not to be confused with, the presence of repeated runout (RRO) error in a PES over a complete revolution of the disc. This latter phenomenon occurs due to factors such as disc eccentricity (i.e., a slight disc shift after the servo data are read) and slight radial errors in the location of the various servo data fields that deviate from a true circle about the rotational axis.
The presence of RRO in a PES can be detected by holding a selected head in a fixed radial relationship to a track on the corresponding disc. If the track is perfectly circular, there will be no variation in detected intra-track position from each of the servo data fields encountered as the disc makes a complete revolution. On the other hand, the presence of RRO will be exhibited as a frequency signal superimposed onto the PES that repeats over each disc revolution. Even though the head is maintained perfectly still over the disc, the RRO will cause the head to appear to wobble or oscillate about the selected track.
Thus, efforts to reduce RRO advantageously correct track eccentricities, but do not otherwise remove nonlinearities in the PES as the head is advanced across the radial width of a selected track. For reference, efforts to compensate for such RRO are exemplified by U.S. Pat. No. 5,539,714 issued to Andrews, Jr. and U.S. Pat. No. 5,978,169 issued to Woods, each of which generally disclose generating and injecting a feed-forward correction signal into the servo loop in relation to the error in the radial placement of each set of servo position data about the track.
A continued trend in the industry is to provide disc drives with ever greater levels of data storage and data rate capabilities, including ever greater track densities. Thus, there remains a continued need for improvements in the art to compensate for PES nonlinearities.
SUMMARY OF THE INVENTION
In accordance with preferred embodiments, a servo circuit of a disc drive data handling system operates to generate a sequence of position error signal (PES) samples indicative of head position with respect to a selected track on a disc surface using servo position data samples transduced from the selected track. The servo circuit further operates to generate the sequence of PES samples in relation to an initial, first set of compensation values selected to reduce nonlinearities in said PES samples across the radial width of said track. Preferably, the initial set of compensation values are globally applied during manufacturing to a population of nominal disc drives and serve to provide a first cut at PES linearization.
Nonlinearities remaining in the PES samples obtained using the initial compensation values will tend to indicate the extent to which the initial compensation values were deficient in removing all such nonlinearities from the PES samples. Thus, the PES samples obtained during the foregoing operation are combined with the first set of compensation values to generate an updated, second set of compensation values which will tend to exhibit improved linearization performance. In some preferred embodiments, the calibration pro
Heydt Jeffrey A.
Ray James B.
Fellers , Snider, et al.
Hudspeth David
Seagate Technology LLC
Slavitt Mitchell
LandOfFree
Reducing position error signal nonlinearity through... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Reducing position error signal nonlinearity through..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Reducing position error signal nonlinearity through... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3221019