K. L. Wood*, I. Busch-Vishniac*, D. Neikirk+, and W. Weldon*+

Departments of Mechanical Engineering* and Electrical Engineering+

The University of Texas at Austin, ETC 5.160

Austin, TX 78712-1063

(512) 471-0095 FAX: (512) 471-8727

Microdynamical systems have been studied for a number of years, with concentration on sensors and, more recently, actuators. Very little work has been done on integrating microdynamical components into systems that satisfy mechanical tasks on macroscopic scales. In this work we are studying microdynamical components with appropriate controlling hardware to produce a surface which is actively deformable. Active surfaces can operate on global or local surface characteristics. In global surface control, an overall surface structure is prescribed and can change in time, as in an active airfoil where the lift must change over time. In local surface control, sensing, control and actuation all are accomplished locally in order to produce either a purely local or a global effect. For example, local control could be used to minimize local pressure gradients across a surface or to produce a global wave which propagates along the surface. In local control, the surface could be composed of cellular automata, where all necessary operations are carried out within each individual cell or a cell and its nearest neighbors, thus avoiding the difficulties of global interconnections.

Because of the potential advantages of local surface control, this work concentrates on creating micron-scale surface deformations of critical machine elements to produce macro-scale system responses. In particular, our focus is on the design and demonstration of smart journal and thrust bearings capable of using embedded sensors and actuators to dynamically change the surface geometry of the bearing (Figure 1). The ability to actively deform bearing surfaces allows for the design of bearings which are less prone to failure, the design of bearings with greater load carrying abilities, and a fundamental study of the effect of surface geometries and fluid conditions on bearing performance, such as start-up and shut-down conditions.

At this juncture in the research, we have chosen silicon membrane structures for the actuators and sensors, with in-plane dimensions ranging from 50x50 square microns to 1000x1000 square microns and out-of-plane deflections from 0.1-50 microns (Figure 2). Five research tasks are being implemented to support the development of the membrane elements: bearing modeling, sensor and actuator modeling and experiments, bearing experiments, generator modeling and experiments, and system integration (system design). Specifically, bearing modeling, with integrated fluid dynamics and membrane solid mechanics, has resulted in a robust set of feasible membrane and static surface configurations. New scaling-law relationships in surface array structures have demonstrated (from simulations and experimentation) approximately 150% increases in local bearing pressures and 30-100% increases in load carrying capacity. In addition, Fabry-Perot pressure sensors and optical interconnects have been fabricated. These structures, a pneumatic manifold, an oil-film bearing surface, and an adapted precision grinding machine form the baseline experimental environment for the smart bearings (Fig. 3). Experimental confirmation of the fluid dynamics modeling is under way, and preliminary results indicate predictions of bearing load capacity improvement are correct. We are also considering manufacturing techniques for use in true journal bearings (i.e., cylindrical geometires).


This work is supported by the ARPA Embedded Microsystems Program.



Figure 1: Example bearing system, showing salient features, and linear slider bearing model.


Figure 2: Etched silicon bearing pad.


Figure 3: Thrust bearing experimental testbed showing Harig grinder and thrust bearing subassembly.