Leakage Flow Analysis for a MEMS Rotary Engine

Abstract

An internal leakage flow analysis is presented for a MEMS fabricated rotary engine in order to establish design parameters for micro engine sealing systems. This research is part of the MEMS Rotary Engine Power System (REPS) group effort to develop a portable power system based on an integrated generator and Wankel rotary internal combustion engine. In order to have acceptable system efficiency, it is necessary to suppress internal leakage and thereby maintain a critical level of compression ratio. There are two inherent leakage paths in rotary engines, which result in blowby and reduced compression ratio: leakage around the apexes of the rotor and leakage across the rotor faces. These sealing issues arise due to the large pressure gradients, which occur along these leakage paths in the combustion chamber. It is the aim of this work to examine the effects of reduced scale on both traditional and novel rotary engine apex sealing mechanisms. In contrast to the macro scale, viscous forces have an increased importance in micro scale engines since Re~.01. A simplified Poiseuille-Couette flow model has been developed to analyze the leakage flows of rotary type engines. Since the Reynolds number for the MEMS REPS is extremely small, the model assumes that the flow is laminar, viscous, incompressible, and steady with air as the working fluid. The model indicates that if a 1 μm gap can be maintained between the housing and moving parts (rotor apexes and faces), leakage flows at expected engine operation speeds will only reduce the compression ratio from 8.3:1 to 6.1:1 so long as the rotation speed is greater than 10,000 rpm. It is doubtful that a traditional or simple micromachine design will yield such a gap and therefore several novel, integrated sealing approaches are under investigation. The model will determine design specification for one of these approaches, an integrated cantilever flexure apex. In conjunction with the theoretical model, a scaled engine experiment at the macro scale is used to verify the modeling effort. The scaling of the experiment complies with Reynolds scaling and ensures that Hele-Shaw flow within the leakage paths is maintained. The experiment does not operate as a functional engine, rather the experiment is designed to maintain a precise clearance between the rotor and housing. In order to preclude additional pressure driven flow effects, an electric motor is used to spin the rotor and simulate the rotation expected due to the combustion pressure acting on the rotor face.