Artificial neural network application for magnetic core width prediction and modelling

Abstract

Increased demands for higher storage capacity solution have driven the Hard Disk Drive (HDD) technological boundaries. As the Perpendicular Magnetic Recording (PMR) head shows promising increase in Areal Density away from the limit of Longitudinal Magnetic Recording, HDD companies have switch to 100% PMR drives. PMR heads requires tight physical specifications fabricating its Writer Element in order control the magnetic flux footprint of the writer on the disk. This magnetic footprint is also called the MCW (Magnetic Core Width). MCW variations in PMR head results to significant yield loss in DET (Dynamic Electrical Test). In addition to that, continuous tweaking in Wafer and Slider Fab process to improve yield contributes to changes in MCW performance during DET. A new method that will learn and predict the MCW model accurately is thus necessary to successfully control MCW variation. An Artificial Neural Network Multilayer Perceptron architecture was developed and used to derive the MCW model from Wafer & Slider process parameters. The Artificial Neural Network model was compared with conventional Multiple Linear Regression (MLR) method and has shown that ANN gives better accuracy in predicting the final MCW than MLR by 30%.