Browse Prior Art Database

Statistical Model of Disk Pack Particle Count Behavior

IP.com Disclosure Number: IPCOM000063313D
Original Publication Date: 1985-Mar-01
Included in the Prior Art Database: 2005-Feb-18

Publishing Venue

IBM

Related People

Authors:
Prater, WL [+details]

Abstract

A statistical algorithm has been developed to model and predict the particle count behavior in magnetic recording disk packs. By using four to eight minutes of test data, the algorithm is able to discriminate between disk packs that will clean up and pass the particle count specification within twelve minutes and disk packs that will remain dirty and fail the specification after twelve minutes. Product cleanliness is required at all test operations of the magnetic recording disk file process line to prevent head crashes of the head-disk assembly in field operation. Following the assembly of the magnetic disks into the bearing housing, one of the test sequences is the measurement of airborne particle contamination in the disk pack by a particle counter.