Numerical study of the flow-structure interactions in an air- or helium-filled simulated hard disk drive

Abstract

Since its invention, the Hard Disk Drive (HDD) has been the most widely-used device for data storage. Recently, the volume of data is getting larger and the corresponding rotation speed of the HDD is increasing to allow for better data transfer. The decreasing size of the disk is increasing the density of data on the disk surface. As a result, the positioning accuracy of the Suspension Slider Unit (SSU), where the magnetic head is mounted, is the problem that has to be overcome for better performance of the HDD. Additionally, the increased rotating speed of the disk induces unsteady flow between each pair of disks. This unsteady flow becomes turbulent around the SSU and induces vibrations on the SSU which deteriorate the performance of the HDD. There have been many investigations to understand the fluid mechanics phenomena inside the HDD filled with air. Additionally, many modifications have been tried to minimize the flow-induced vibration on the SSU by placing a blockage upstream of the arm to generate a low velocity region. However, none of these investigations have explored the effect of using gases other than air. In this work, the flow physics in the HDD is investigated numerically with the drive filled with air or helium. Numerical analyses were performed using the commercial code (ANSYS/CFX) with an expanded 2 × model simulating Seagate cheetah 2.5-inch drive. Despite obvious un-addressed issues in sealing the HDD, the unsteady characteristics of the flow are dissipated sufficiently faster in helium than in air so as to warrant further studies addressing the more practical issues of working with helium. Of particular importance is the unsteady flow around the SSU. This leads to lower levels of flow-induced vibration in the case of helium flow. As such, HDD performance may be improved by using helium to improve the dynamics of the HDD at higher rotation speeds. For both air- and helium-filled drives, calculations have been performed with two different locations of the SSU and two different angular velocities, 1,000 and 3,000 rpm corresponding to 5,000 and 15,000 rpm in 3.5-inch commercial drive. Not only is it shown that the helium-filled drive suffers lower positioning errors, but also the underlying flow physics responsible for such improvement are explained.