Numerical Investigation of the Flow Topology in a Bidirectional Vortex Engine
Type of DegreePhD Dissertation
MetadataShow full item record
The characterization of swirl-driven phenomena, including cyclonic flows, continues to receive attention in the propulsion community. This may be attributed to the beneficial properties of swirl when integrated into several classes of vortex engines. Moreover, the use of computational techniques in the treatment of cyclonic flows has become justified by the progressive evolution of hardware and the affordability of computing resources relative to experimentation. Bearing these developments in mind, the present investigation will focus on the use of a finite-volume computational fluid dynamics (CFD) solver to characterize the basic nature of the wall-bounded cyclonic motion in a swirl-driven thrust chamber known as the “Bidirectional Vortex Engine.” In this context, a non-reactive, cold-flow simulation is performed using an idealized configuration of a right-cylindrical chamber. The latter comprises eight tangential injectors through which air is introduced under both steady and transient flow conditions. The flow is initially considered to be incompressible and inviscid to help verify existing analytical models. At a later stage, the study is modified to include simulations under both laminar and turbulent conditions. To minimize cell skewness around injectors, a fine tetrahedral mesh is constructed and then converted into polyhedral cells, namely, to improve convergence characteristics and precision. Once convergence is achieved in all variables, our principal variables are evaluated and compared to existing formulations that are often referred to as Complex-Lamellar, Beltramian, and Constant Shear Stress core flow models. At the outset, one is able to (a) characterize the wall-bounded cyclonic motion under inviscid, laminar, and turbulent conditions; (b) determine the appropriate turbulence models that will accurately predict the underlying anisotropic behavior of the flow; and (c) help to identify the threshold properties, expressed as functions of non-dimensional parameters, that will either initiate the inception or trigger the breakdown of fully-developed cyclonic motions.