Optimization of hard disk drive components

Computer hard disk drives have been used as large capacity storage devices for better than three decades. Over the course of that time, the amount and density of information held has increased manyfold. Current hard disk drives combine technology from the fields of magnetics, material science, signal processing, mechanics and fluid dynamics. Optimization allows a number of the considerations from these various fields to be addressed simultaneously.; This dissertation focuses on the development and application of global optimization techniques to the problem of hard disk drive component design. Genetic algorithms and the Simulated Annealing (SA) algorithm are compared for their applicability to this optimization problem. Through empirical and theoretical study SA has been found to be superior for this problem. An accompanying software package that utilizes an implementation of SA has been developed that allows for the multi-objective optimization of air bearing slider and suspension designs.; In order to understand the underlying physics of the problem, a description of the governing equation of the physical process--the Reynolds' equation--is presented in addition to the method implemented to solve it numerically. Also, a survey of deterministic and stochastic global optimization techniques is given to provide a backdrop for the current study.; The parameters used in the simulated annealing algorithm that allow it to achieve global optima for the problem at hand in a more efficient manner are determined. Separate suggestions are given for the slider design and suspension design problems. The suspension design problem demands an implementation that requires more function evaluations due to the sensitivity of the objective function to the parameters employed. Several examples are given in support of this claim.; Application of SA to the problem of slider altitude sensitivity optimization is presented. It is demonstrated that altitude insensitive designs can, indeed, be achieved through use of SA. The application of SA to the problem of sub-ambient pressure minimization is also presented. It is demonstrated that sub-ambient pressure minimization can be achieved through use of SA.; The problem of slider dynamic fly height variation is addressed through optimization. Through modification of static characteristics of the bearing, an attempt was made to modify the dynamic characteristics of the bearing. Although the optimization algorithm optimized the quantities that were included in the objective function, the resulting designs did not minimize the dynamic fly height variation.; Finally, the application of SA to a specific example of suspension optimization is shown. It demonstrates that the procedure of structural optimization is possible through application of the SA algorithm using an appropriately generous cooling schedule.