A soft error rate model for predicting off-track performance

Abstract

Specifying the head and determining the track density are critical, interdependent decisions made early in the development of a hard disk drive. We have created a soft error rate model to be used as a tool to aid in making these decisions. We have also developed methods to estimate the side reading and writing parameters of a head using well known spinstand tests. In addition to these four reading and writing parameters, the model requires an estimate of the track misregistration (TMR) and a sixth input parameter, the effective on-track signal-to-noise ratio (SNR/sub eff/) of the channel. The utility of the model is demonstrated in a test case based on a system with a thin film head/medium and a peak detection channel. The model is used to predict error rate both as a function of track density for a given head and as a function of head width for a given track density. The validity of the model is demonstrated by the close agreement of simulated 747 curves generated by the model with experimental 747 curves measured as a spinstand. >