Session L30: Compressible Flow: Special Topics and Applications

Control of a Normal Shock Boundary Layer Interaction with Ramped Vanes of Various Sizes

A novel vortex generator design positioned upstream of a normal shock and a subsequent diffuser was investigated using large eddy simulations. In particular, ``ramped-vane'' flow control devices with three difference heights relative to the incoming boundary layer thickness (0.34$\delta $ 0.52$\delta $ and 0.75$\delta $ were placed in a supersonic boundary layer with a freestream Mach number of 1.3 and a Reynolds number of 2,400 based on momentum thickness. These devices are similar to subsonic vanes but are designed to be more mechanically robust while having low wave drag. The devices generated strong streamwise vortices that entrained high momentum fluid to the near-wall region and increased turbulent mixing. The devices also decreased shock-induced flow separation, which resulted in a higher downstream skin friction in the diffuser. In general, the largest ramped-vane (0.75$\delta )$ produced the largest reductions in flow separation, shape factor and overall unsteadiness. However, the medium-sized ramped vane (0.52$\delta )$ was able to also reduce both the separation area and the diffuser displacement thickness. The smallest device (0.34$\delta )$ had a weak impact of the flow in the diffuser, though a 10{\%} reduction in the shape factor was achieved.   

Second-mode control in hypersonic boundary layers over assigned complex wall impedance

The durability and aerodynamic performance of hypersonic vehicles greatly relies on the ability to delay transition to turbulence. Passive aerodynamic flow control devices such as porous acoustic absorbers are a very attractive means to damp ultrasonic second-mode waves, which govern transition in hypersonic boundary layers under idealized flow conditions (smooth walls, slender geometries, small angles of attack). The talk will discuss numerical simulations modeling such absorbers via the time-domain impedance boundary condition (TD-IBC) approach by Scalo et al. \textit{Phys. Fluids} (2015) in a hypersonic boundary layer flow over a 7-degree wedge at freestream Mach numbers $M_{\infty}$ $=$ 7.3 and Reynolds numbers Re$_{\mathrm{m}}=$ $1.46 \cdot 10^6$ . A three-parameter impedance model tuned to the second-mode waves is tested first with varying resistance, R, and damping ratio, $\zeta$, revealing complete mode attenuation for R \textless \ 20. A realistic IBC is then employed, derived via an inverse Helmholtz solver analysis (Patel et al. AIAA 2017-0460) of an ultrasonically absorbing carbon-fiber-reinforced carbon ceramic sample used in recent hypersonic transition experiments by Dr. Wagner and co-workers at DLR-G\"{o}ttingen.   

Evolution of solenoidal and dilatational perturbations in transitional supersonic and hypersonic boundary layers.

We have investigated the interaction between the dilatational and solenoidal components of instability waves relying on DNS simulations of temporally-evolving compressible boundary layers ranging from Mach numbers of 2.0 to 10.0. For idealized flow conditions at subsonic-to-moderate supersonic speeds, transition to turbulence occurs due to amplification of Tollmien-Schlichting (T-S) waves (first Mack mode) exponentially amplified until nonlinear breakdown and transition to turbulence occurs. Under the same conditions, at hypersonic speeds, transition is governed by acoustically resonating trapped waves (second Mack mode). While the former are expected to be solenoidal in nature and the latter predominantly dilatational, we demonstrate that, in general, they always coexist and that, even at Mach=10 there is an appreciable energy transfer from the dilatational to the solenoidal at limit-cycle amplitude conditions in 2D simulations. In three-dimensional simulations very rapid breakdown is observed. Mechanisms of energy exchange between the dilatational and solenoidal components during the transition will be discussed.   

Simulations of Wakes and Parachute Environments for Supersonic Flight Test Design.

NASA's ASPIRE (Advanced Supersonic Parachute Inflation Research and Experiments) project is a risk-reduction activity for a future mission, Mars2020. ASPIRE will investigate the supersonic deployment, inflation and aerodynamics of a full-scale disk-gap-band (DGB) parachute in the wake of a slender body at high altitudes over Earth. The leading slender body has about 1/6-th the diameter of the entry capsule that will use this parachute for descent at Mars. ASPIRE flight test design (targeting, safety and recovery) requires models for deployment, inflation and aerodynamic performance of the parachute. However, there is limited flight and experimental data for supersonic DGBs behind slender bodies. This presentation describes the use of CFD in supplementing the available data to construct a parachute aerodynamics model for ASPIRE. Simulations are used to understand the effects of the leading body on the wake, and on the canopy loads, results of which will be presented. The first flight test is scheduled for September 2017. Comparisons of preliminary test data against the pre-test parachute model will be presented.   

Modeling of thermodynamic non-equilibrium flows around cylinders and in channels

Numerical simulations for two different types of flash-boiling flows, namely shear flow (flow through a de-Laval nozzle) and free shear flow (flow past a cylinder) are carried out in the present study. The Homogenous Relaxation Model (HRM ) is used to model the thermodynamic non-equilibrium process. It was observed that the vaporization of the fluid stream, which was initially maintained at a sub-cooled state, originates at the nozzle throat. This is because the fluid accelerates at the vena-contracta and subsequently the pressure falls below the saturation vapor pressure, generating a two-phase mixture in the diverging section of the nozzle. The mass flow rate at the nozzle was found to decrease with the increase in fluid inlet temperature. A similar phenomenon also occurs for the free shear case due to boundary layer separation, causing a drop in pressure behind the cylinder. The mass fraction of vapor is maximum at rear end of the cylinder, where the size of the wake is highest. As the back pressure is reduced, severe flashing behavior was observed. The numerical simulations were validated against available experimental data.   

Simulations of supersonic highly under-expanded hydrogen jets

The pressure drop across choke valves required to transport natural gas can be in the order of several hundred bars, leading to the development of supersonic under-expanded jets. When considering a real gas, the gas can cool upon expansion, a phenomenon which can be explained by the Joule-Thomson effect. This study compares the effects of using ideal and real gas equations of state, using a computational model in which hydrogen is released from a high-pressure tank, through a converging nozzle, into a chamber containing hydrogen at near-atmospheric conditions. The initial studies were carried out using an ideal gas assumption and nozzle pressure ratios of 10, 30 and 70 and the results were validated against existing literature. To account for the Joule-Thomson effect, ideal and real gas simulations were then carried out with a pressure ratio of 70. For the real gas model, the Peng-Robinson equation of state was chosen. At the nozzle exit, the ideal gas model underestimates the velocity and overestimates the temperature and density; as the flow expands, the flow properties are the same up to the Mach disk, at which point the ideal gas underestimates the Mach number and predicts a higher temperature and density than the Peng-Robinson model due to the absence of cooling.   

Comparison of Themodynamic and Transport Property Models for Computing Equilibrium High Enthalpy Flows

To study the flow of high temperature air in vibrational and chemical equilibrium, accurate models for thermodynamic state and transport phenomena are required. In the present work, the performance of a state equation model and two mixing rules for determining equilibrium air thermodynamic and transport properties are compared with that of curve fits. The thermodynamic state model considers 11 species which computes flow chemistry by an iterative process and the mixing rules considered for viscosity are Wilke and Armaly-Sutton. The curve fits of Srinivasan, which are based on Grabau type transition functions, are chosen for comparison. A two-dimensional Navier-Stokes solver is developed to simulate high enthalpy flows with numerical fluxes computed by AUSM+-up. The accuracy of state equation model and curve fits for thermodynamic properties is determined using hypersonic inviscid flow over a circular cylinder. The performance of mixing rules and curve fits for viscosity are compared using hypersonic laminar boundary layer prediction on a flat plate. It is observed that steady state solutions from state equation model and curve fits match with each other. Though curve fits are significantly faster the state equation model is more general and can be adapted to any flow composition.   

Investigation of Cooling Water Injection into Supersonic Rocket Engine Exhaust

Water spray cooling of the exhaust plume from a rocket undergoing static testing is critical in preventing thermal wear of the test stand structure, and suppressing the acoustic noise signature. A scaled test facility has been developed that utilizes non-intrusive diagnostic techniques including Focusing Color Schlieren (FCS) and Phase Doppler Particle Anemometry (PDPA) to examine the interaction of a pressure-fed water jet with a supersonic flow of compressed air. FCS is used to visually assess the interaction of the water jet with the strong density gradients in the supersonic air flow. PDPA is used in conjunction to gain statistical information regarding water droplet size and velocity as the jet is broken up. Measurement results, along with numerical simulations and jet penetration models are used to explain the observed phenomena. Following the cold flow testing campaign a scaled hybrid rocket engine will be constructed to continue tests in a combusting flow environment similar to that generated by the rocket engines tested at NASA facilities.   

Examination of rapid phase change in copper wires to improve material models and understanding of burst

Electrically-pulsed wires undergo multiple phase changes including a postulated metastable phase resulting in explosive wire growth. Simulations using the MHD approximation attempt to account for the governing physics, but lack the material properties (equations-of-state and electrical conductivity) to accurately predict the phase evolution of the exploding (bursting) wire. To explore the dynamics of an exploding copper wire (in water), we employ a digital micro-Schlieren streak photography technique. This imaging quantifies wire expansion and shock waves emitted from the wire during phase changes. Using differential voltage probes, a Rogowski coil, and timing fiducials, the phase change of the wire is aligned with electrical power and energy deposition. Time-correlated electrical diagnostics and imaging allow for detailed validation of MHD simulations, comparing observed phases with phase change details found in the material property descriptions. In addition to streak imaging, a long exposure image is taken to capture axial striations along the length of the wire. These images are used to compare with results from 3D MHD simulations which propose that these perturbations impact the rate of wire expansion and temporal change in phases. If successful, the experimental data will identify areas for improvement in the material property models, and modeling results will provide insight into the details of phase change in the wire with correlation to variations in the electrical signals.   

Supersonic nozzle flow and condensation analysis by the non-equilibrium molecular dynamics

In this research, we study supersonic inviscid flow through a Laval nozzle by means of the non-equilibrium molecular dynamics, and demonstrate the capability of the method for analysis of the nozzle flow and condensation. The flow is driven by the pressure difference imposed by heat baths attached to both ends of the nozzle. Only the repulsive force is applied between molecules and the nozzle wall to minimize the viscous effect. With Argon gas (Lennard-Jones potential), the flow properties follow the isentropic expansion except on the nozzle end. Water vapor flow is also investigated with the modified Lennard-Jones potential to observe the condensation of the water through the isentropic expansion. The result shows the nucleation and growth of molecule clusters in the region after the nozzle throat.   

Transonic flow of steam with non-equilibrium and homogenous condensation

A small-disturbance model for studying the physical behavior of a steady transonic flow of steam with non-equilibrium and homogeneous condensation around a thin airfoil is derived. The steam thermodynamic behavior is described by van der Waals equation of state. The water condensation rate is calculated according to classical nucleation and droplet growth models. The current study is based on an asymptotic analysis of the fluid flow and condensation equations and boundary conditions in terms of the small thickness of the airfoil, small angle of attack, closeness of upstream flow Mach number to unity and small amount of condensate. The asymptotic analysis gives the similarity parameters that govern the problem. The flow field may be described by a non-homogeneous transonic small-disturbance equation coupled with a set of four ordinary differential equations for the calculation of the condensate mass fraction. An iterative numerical scheme which combines Murman {\&} Cole's (1971) method with Simpson's integration rule is applied to solve the coupled system of equations. The model is used to study the effects of energy release from condensation on the aerodynamic performance of airfoils operating at high pressures and temperatures and near the vapor-liquid saturation conditions.