Image processing techniques for analysis of air bearing surfaces

Numerous studies have shown the importance of surface parameters (i.e., camber, crown, and twist) on air bearing performance. These parameters are used as both modeling inputs and manufacturing controls. Measured values are, however, sensitive to both the assumed shape of the surface (e.g., biquadratic) and the numerical procedures used to extract them from the data (e.g., least squares regression). This sensitivity is becoming even more important as fly heights are reduced below 10 nm. The purpose of this paper is to present the numerical algorithms required to take raw WYKO or ZYGO data of an air bearing surface and automatically convert it into a format that can be used to determine surface parameters or can be directly incorporated into air bearing simulations. The former is useful for determining more accurate representations of the surface, while the latter is useful for determining the impact of these surface approximations.