Numerical Examination of Flow Field Characteristics and Fabri Choking of 2D Supersonic Ejectors

Author

Brett G. Morham, California Polytechnic State University, San Luis ObispoFollow

DOI: https://doi.org/10.15368/theses.2010.107
Available at: https://digitalcommons.calpoly.edu/theses/340

Date of Award

6-2010

Degree Name

MS in Aerospace Engineering

Department/Program

Aerospace Engineering

Advisor

Dianne Deturris

Abstract

An automated computer simulation of the two-dimensional planar Cal Poly Supersonic Ejector test rig is developed. The purpose of the simulation is to identify the operating conditions which produce the saturated, Fabri choke and Fabri block aerodynamic flow patterns. The effect of primary to secondary stagnation pressure ratio on the efficiency of the ejector operation is measured using the entrainment ratio which is the secondary to primary mass flow ratio.

The primary flow of the ejector is supersonic and the secondary (entrained) stream enters the ejector at various velocities at or below Mach 1. The primary and secondary streams are both composed of air. The primary plume boundary and properties are solved using the Method of Characteristics. The properties within the secondary stream are found using isentropic relations along with stagnation conditions and the shape of the primary plume. The solutions of the primary and secondary streams iterate on a pressure distribution of the secondary stream until a converged solution is attained. Viscous forces and thermo-chemical reactions are not considered.

For the given geometry the saturated flow pattern is found to occur below stagnation pressure ratios of 74. The secondary flow of the ejector becomes blocked by the primary plume above pressure ratios of 230. The Fabri choke case exists between pressure ratios of 74 and 230, achieving optimal operation at the transition from saturated to Fabri choked flow, near the pressure ratio of 74. The case of optimal expansion yields an entrainment ratio of 0.17. The entrainment ratio results of the Cal Poly Supersonic Ejector simulation have an average error of 3.67% relative to experimental data. The accuracy of this inviscid simulation suggests ejector operation in this regime is governed by pressure gradient rather than viscous effects.