Browse Prior Art Database

Mini bias test - line screen and predictive failure analysis Disclosure Number: IPCOM000013656D
Original Publication Date: 2000-Feb-01
Included in the Prior Art Database: 2003-Jun-18

Publishing Venue



Mini bias test line screen and predictive failure analysis A method is described to detect defective pivot bearings in disk drives by evaluating actuator coil current anomalies during track following. The technique is used to prevent disk drive failures due to problems with the actuator’s pivot bearing by using the disclosed method either as a screen at the disk drive manufacturing line or as a predictive failure analysis tool to prevent data loss from occurring at the customer’s application. When an actuator is track following, the coil current is continually adjusted to move the head a fraction of a track to maintain it following along the track centerline. Ideally, for a nominal actuator, the coil current should be linear with no discontinuities over the entire data band. It would be expected that coil current requirements for adjacent tracks should be nearly the same. Windage force coming from the rotating spindle, and flex cable force created by bending of the flex cable are generally non-zero and both must be overcome by the actuator VCM during track following. They combine to give a plot of track position versus coil current with good linearity and a negative slope. Torsional resistance in the bearing coming from elastic deformation of the balls and raceways and viscous shearing of the grease combine to give equal resistance to both the inbound and outbound seek directions. This shows up as a torque separation between the inbound and outbound plots, also known as hysteresis (H). Other good indicators of the linearity of the pivot bearing torque behavior are: the bias current correlation coefficient (r), and the least squares linear curve fit (Y A mX) of bias current with respect to cylinder. To screen out defective drive an individual disk drive’s statistics can be compared to population statistics.