An attempt has been made to investigate experimentally the
effects of flow expansions caused by moving expansion waves
on isotropic turbulence. The flow field was generated by
using a variety of rectangular grids of different mesh size.
The Reynolds number of the flow based on the mesh size,
ranged from 50,000 to 200,000 while the turbulent Reynolds
number, based on Taylor's microscale was between 150 and
650. Mild interactions with expansion ratios of the order of
1.8 were achieved. The ratio of the eddy turnover time to
that of the strain rate of the interaction reached values of
the order of 1000. A custom designed vorticity probe was
used to measure the rate-of-strain, the rate-of-rotation and
the velocity-gradient tensors. Although the strength of the
generated expansion waves was mild, 50 to100 s-1, its effect
on damping fluctuations of turbulence was clear. Vorticity
fluctuations were reduced dramatically more than velocity or
pressure fluctuations. Attenuation of longitudinal velocity
fluctuations has also been observed. It appears that the
attenuation increases in interactions with higher Reynolds
number. Moreover, attenuation occurs at large distances from
the grid where length scales of the incoming flow are high
and turbulence intensities are low. Thus large in size
eddies with low velocity fluctuations are affected the most
in these interactions.
[MN.002] Reshock of a Convergent, Unstable System
Steven Batha, N. D. Delamater, J. R. Fincke, R. M. Hueckstaedt, N. E. Lanier, G. R. Magelssen, J. M. Taccetti (Los Alamos National Laboratory), K. W. Parker, S. D. Rothman, C. J. Horsfield (AWE, Aldermaston)
The behavior of hydrodynamic instabilities is different in planar and convergent geometries. Implosion of foam-filled cylinders using the Omega laser allows the examination of the effects of convergence on the evolution of the Richtmyer-Meshkov instability. Single-shell implosions have been used to determine the zero-order hydrodynamic motion and the effect of surface roughness on seeding the instability in the compressible, convergent, miscible, strong-shock regime. This experiment has been extended to measure the growth after reshock of a rapidly growing R-Z perturbation. Placing a solid core at the center of the cylinder generates the reshock by reflecting the main shock. Perturbations of wavelength 2.5, 6.0, and 9.0 microns and peak-to-valley amplitude 2.0 micron are machined into the outside of a thin Al marker layer embedded within the cylinder. When subjected to a strong reshock, the compression of the marker layer and subsequent regrowth are easily observed. Comparisons with the Eulerian radiation-hydrodynamics code RAGE are made.
This research was performed by the Los Alamos National
Laboratory under the auspices of the United States
Department of Energy under contract No. W-7405-ENG-36.
[MN.003] Cross density variations and vorticity generation in compressible shears
M. Belan, S. De Ponte (Politecnico di Milano), D. Tordella (Politecnico di Torino)
The study is based on the observation of underexpanded
hypersonic jets and relevant turbulent mixings downstream of
the barrel shok [Belan et al.,2001,19th ICIASF; Belan
et al.,2004,Astr.Space Sci., to appear]. The
stagnation/ambient pressure ratio is 10^3 - 10^4, the
density variations were obtained using different gases in
the jet and in the ambient. By using He, Ar and air, the
stagnation/ambient density ratio was changed by one order of
magnitude while keeping fixed the pressure ratio. Three jets
have been observed: air in air (Ma\sim20), He in air
(Ma\sim30) and Ar in air (Ma\sim30), Ma=Mach number
upstream of the first disk. The near and mid-term jet
evolution was visualized through the electron beam
technique. When the density after the first Mach disk is of
the order of the ambient density, as for the air-air and
Ar-air jets, the spreading of the turbulent mixing which
surrounds the jet core is very small, comparatively smaller
than when the density distribution is lower than that in the
ambient (He jet). This different dynamics cannot be
explained in terms of compressibility effects only. The
changes in the morphology are associated to changes in the
transport properties of the turbulent shears. This is
discussed in the context: i) of the compressible vorticity
equation and of its source terms, ii) since in this
experiment the global Ma and Re numbers are comparable -
of the Sonic Eddy Model [Breidenthal,1992].
[MN.004] DSMC Investigation of Interacting Parallel Supersonic Free Jets
Wenhai Li, Foluso Ladeinde (SUNY Stony Brook, New York, USA)
Our interest in adjacent parallel jets is motivated by the
relevance to the design of rocket propulsion systems. The
direct simulation Monte Carlo (DSMC) method is the preferred
method of analysis of rarefied compressible flows because of
better accuracy compared to continuum prediction methods. In
this paper, we apply the DSMC method to investigate the
dynamics of three-dimensional flow of two interacting,
parallel, supersonic free jets, including the resolution of
the internal structure of the shock system. The jets come
from two parallel orifices and exhaust into a low-pressure
chamber. Bird's no-timing-counter scheme is used for the
intermolecular collisions, while the variable-soft-sphere
(VSS) molecular model is adopted for both monatomic and
diatomic gases. Non-equilibrium effects are considered for
the diatomic gases. The effects of the geometric arrangement
of the orifices and the boundary pressure conditions are
investigated and will be reported, as will the effects of
the roughness of the cell network and the number of
molecules in a cell.
[MN.005] Planar Velocimetry of a Supersonic Jet in Subsonic Compressible Crossflow
Steven Beresh, John Henfling, Rocky Erven, Russell Spillers (Sandia National Laboratories)
A stereoscopic particle image velocimetry (PIV) instrument
has been constructed for a transonic wind tunnel to study
the interaction created by a supersonic axisymmetric jet
exhausting from a flat plate into a subsonic compressible
crossflow. Data have been acquired in the crossplane of the
interaction at a single station in the farfield, in which
the bulk particle motion is aligned with the out-of-plane
velocity component. The resulting vector fields distinctly
show the strength and location of the induced
counter-rotating vortex pair as well as the remnant of the
horseshoe vortex that wraps around the jet plume as it first
exhausts from the nozzle. The vortices are visible from the
in-plane vorticity as well as a deficit in the streamwise
velocity component. Data taken for four different values of
the jet-to-freestream dynamic pressure ratio reveal that the
vortex strength, size, and distance from the wall all
increase with jet pressure. An uncertainty analysis also is
provided.
[MN.006] Prediction of a jet in supersonic crossflow
C. Randall Truman, Amol Palekar, Peter Vorobieff (U. New Mexico)
Numerical simulation of a circular jet injected transversely
into a supersonic crossflow at Mach 2 is described.
Penetration and mixing characteristics are compared with
experiments by Gruber et al. where several different
injected fluids were used to vary the momentum ratio as well
as the effect of compressibility. This work is in support of
improved predictions of mixing in chemical laser nozzles
with injection.
[MN.007] DNS of a shock wave turbulent boundary layer interaction at M=2.25
Sergio Pirozzoli, Francesco Grasso (University of Rome `La Sapienza'), Thomas B. Gatski (NASA Langley Research Center)
The flow field associated to the impingment of an oblique
shock wave on a supersonic turbulent boundary layer
developing on a flat plate at Mach 2.25 is analyzed by means
of a full direct numerical simulation. For the selected test
conditions significant separation of the turbulent boundary
layer occurs, and the flow exhibits large scale
unsteadiness. The numerical procedure used is a conservative
finite difference discretization that uses a seventh-order
accurate weighted-eno scheme for the inviscid flux, a
fourth-order compact difference approximation for the
viscous flux, and a fourth-order explicit Runge Kutta time
integration algorithm. The results of the DNS are analyzed
with the objective to i) assess the scaling properties of
the mean flow; ii) get some insight into the mechanisms
related to the unsteady oscillation of the shock wave and
the recirculation bubble; and iii) investigate the
controlling mechanisms for turbulence production,
dissipation and transport.
[MN.008] Direct Numerical Simulation of Shockwave and Turbulent Boundary Layer Interactions
Minwei Wu, M. Pino Martin (Princeton University)
Direct numerical simulations of two canonical
configurations, a 24-degree compression ramp and shock
impinging a wall with reflection, are performed to study
shockwave and turbulent boundary layer interactions. The
Mach number of the incoming boundary layers is 2.9 and the
Reynolds number is 2400 based on the momentum thickness. An
assessment of the size and resolution of the computational
domains are given. An analysis of the simulation data are
presented. The shock unsteadiness, turbulence amplification,
and changes of length scales through the shockwave and
turbulent boundary layer interaction are studied.
[MN.009] Analysis of catalysis effects for orbital reentry vehicles
Stefano Bisceglia, Francesco Grasso (University of Rome `La Sapienza'), Giuliano Ranuzzi (CIRA, Capua, Italy)
An analisys of the hypersonic flow in thermochemical
nonequilibrium around the forebody of the Orbital Reentry
Experiment (OREX) is presented for typical reentry
conditions. The numerical approach relies on a k-\epsilon
turbulence model that accounts for the coupling of
turbulence with chemistry. The numerical fluxes are computed
with a 2nd order TVD scheme incorporating finite-rate
chemistry with costant wall temperature and finite-rate wall
catalysis by means of Scott's model. The computed flow
fields are analyzed by considering the most relevant flow
properties and comparing grid converged solutions with
stagnation heat flux data. In order to investigate the
effect of molecule production due to recombination and its
contribution to catalytic process, numerical studies with
both fully catalytic and non-catalytic wall boundary
conditions have been carried out.
[MN.010] Experiments and Computation for a Supersonic Nonequlibrium MHD Channel
Sivaram Gogineni (Innovative Scientific Solutions, Inc.), Igor Adamovich (The Ohio State University), Xiaolin Zhong (University of California, Los Angeles), Roger Kimmel (Air Force Research Laboratory)
This paper summarizes recent developments in on-going work to control transition and turbulence in supersonic boundary layers using magnetohydrodynamic forces on nonequlibrium plasma discharges. Experimental results demonstrate the Lorentz force effect on the boundary layer in M=3 flow of nitrogen ionized by a high-power transverse RF discharge (up to 1 kW) in the presence of a B=1.5 T magnetic field. Boundary layer density fluctuation spectra are measured using the Laser Differential Interferometry (LDI) diagnostics. A decelerating Lorentz force applied to the flow produces an increase of the density fluctuation intensity by up to 10-20% (1-2 dB), compared to the accelerating force of the same magnitude applied to the same flow. Direct numerical simulation of the same flow is also reported. Mean flows for the three-dimensional case have been computed, along with mean and harmonically forced flows for the two-dimensional case. The governing equations of the MHD flow, which are the Navier-Stokes equations with the applied electromagnetic force terms, are computed by a third-order upwinded numerical scheme. The magnetic Reynolds number of the flow is small so that the induced magnetic field is neglected. In the presence of a magnetic field the boundary layer profile changes depending on the direction of the field, and the effect of the magnetic field on the boundary layer is less than the effect of Joule heating.