International Business Machines Corporation

Hitachi Global Storage Technologies Netherlands B.v.

Karl Gong

Gordon James Smith

Hal Hjalmar Ottesen

Bernd Lamberts

Mats Anders Engwall

Larry Joe Gilliland

Thomas E Stanley

Roy P Crawford

Michel P Robert

A position error signal amplitude indicative of the distance between an expected position of a magnetic head relative to the track centerline of a magnetic data storage and retrieval system and an actual position of the magnetic head relative to the track centerline is detected and filtered. The signal is filtered to generate a sway mode signal amplitude indicative of an oscillation of the actual position of the magnetic head relative to the track centerline in the frequency range of the filter. The absolute value of the sway mode signal amplitude is then determined. If the absolute value of the sway mode signal exceeds a predetermined threshold value that correlates to a high probability of impending a head-disk crash, a warning signal is propagated. Alternative embodiments similarly predict the possibility of head-disk crash on the basis of the maximum value of several samples of the position error signal and on the basis of the maximum value of several samples of the square of the position error signal amplitude

Method and apparatus for controlling data read/write between a hard disk and a hard disk controller

A position error signal indicative of the distance between an expected position of a magnetic head relative to the track centerline of a magnetic data storage and retrieval system and an actual position of the magnetic head relative to the track centerline is detected. An absolute value of the position error signal is determined and stored. The detection, determination, and storage is repeated over several revolutions of the magnetic data storage and retrieval system. A maximum of the absolute values of the position error signals is then determined and compared to a threshold value. If the maximum of the absolute values of the position error signals exceeds the threshold value, a warning signal is propagated. An alternative embodiment similarly predicts the possibility of head-disk crash on the basis of the maximum value of several samples of the squares of the position error signals.

Method and apparatus for predictive failure analysis technique for head crashes in hard drives using position error signal

A method and apparatus for monitoring track misregistration that is quicker and that is not limited to worst case assumptions. The present invention includes a first memory for accumulating position error signals for a head to produce an accumulated value, a processor for normalizing the accumulated value to produce a normalized result at a predetermined trigger event and a second memory for adding the normalized result therein to produce a running sum; wherein the head is positioned using the running sum. The processor resets the first memory when the normalized result is produced. A counter is provided for incrementing a count value after each position error signal is added to the first memory. The processor normalizes the accumulated value according to the count value of the counter, and the processor resets the counter as a result. In addition, the processor may normalize the accumulated value according to a population mean and a variance associated with the data storage system. The trigger event may include the accumulated error value reaching a predetermined threshold or a servo seek to the next track. The process may be initiated using a self-test command.

Method and apparatus for monitoring track misregistration

A disk drive having an improved method of adjusting the head switch time to improve disk drive performance. In the process of writing or reading long data files, the heads on the actuator are sequentially selected in order to access all the data tracks forming a cylinder in the shortest time. Head offsets are significantly smaller for switching between heads mounted on the same actuator arm than for switching between heads mounted on different actuator arms. Therefore improved disk drive performance is obtained by using two values of head switch time instead of one. The head switch time allocated for switching from a first head on one actuator arm to a second head on a second actuator arm is made longer than the head switch time allocated for switching from a first head on one actuator arm to a second head on the same actuator arm. By allocating longer time for the head switches between different actuator arms, the error flag rate for heads not ready is reduced. The average head switch time is reduced without increasing error rates by separately optimizing the head switch times for same actuator arm head switches and different actuator arm head switches.

Method to adjust head switch time for improved disk drive performance

An integrated bearing and pump assembly is formed by positioning a hydrodynamic pump coaxially with respect to a hydrodynamic bearing. The integrated assembly provides pressurized fluid for use within the device incorporating the bearing. The hydrodynamic pump comprises a secondary hydrodynamic bearing incorporated into the primary bearing assembly. The secondary bearing is optimized for high flow pumping applications rather than the high pressure, zero flow requirements of the primary bearing. When the bearing assembly is rotated, fluid flow is provided.

Growing community of inventors

San Jose, CA, United States of America

Karl Gong