Kelly A. Stephani

An Examination of Trapped Bubbles for Viscous Drag Reduction on Submerged Surfaces

unbroken gas film coating the solid and in the nanobubble or perhaps microbubble coating regime when an air layer is created with superhydrophobic coatings. We examine an intermediate bubble size regime with a trapped-bubble array (TBA) formed in a tap water environment using electrolysis to grow and maintain bubbles in thousands of millimeter-sized holes on a solid surface. We show that even though surface tension is sufficient to stabilize bubbles in a TBA against hydrostatic and shear forces beneath a turbulent boundary layer, no drag reduction is obtained. Drag measurements were acquired over Reynolds numbers based on plate length ranging from 7.210 4 ReL 3.110 5 using either a force balance for plates mounted in a vertical orientation, or by performing a momentum integral balance using a wake survey for a flat plate mounted in either vertical or horizontal orientation. In that the drag forces were small, emphasis was placed on minimizing experimental uncertainty. For comparison, the flow over a flat plate covered on one side by a large uninterrupted gas film was examined and found to produce large drag reductions of up to 32%. DOI: 10.1115/1.4001273

Detailed modeling and analysis of spacecraft plume/ionosphere interactions in low Earth orbit

Detailed direct simulation Monte Carlo/particle-in-cell simulations involving the interaction of spacecraft thruster plumes with the rarefied ambient ionosphere are presented for steady thruster firings in low Earth orbit (LEO). A nominal mass flow rate is used to prescribe the rocket exit conditions of a neutral propellant species for use in the simulations. The charge exchange interactions of the steady plume with the rarefied ionosphere are modeled using a direct simulation Monte Carlo/particle-in-cell methodology, allowing for a detailed assessment of nonequilibrium collisional and plasma-related phenomena relevant for these conditions. Results are presented for both ram- and wake-flow configurations, in which the thrusters are firing into (ram) or in the direction of (wake) the free stream ionosphere flow in LEO. The influence of the Earths magnetic field on the development of the ion plume is also examined for three different field strengths: two limiting cases in which B→0 and B→∞, and the LEO case in which B=0.5 Gs. The magnetic field is found to have a substantial impact on the resulting neutral and ion plumes, and the gyroscopic motion of the magnetized ions results in a broadening of the ion energy distribution functions. The magnetic field model also incorporates a cross-field diffusion mechanism which is shown to increase the current density sampled far from the thruster.

DNS Study of Transient Disturbance Growth and Bypass Transition Due to Realistic Roughness

Direct numerical simulation was used to investigate the detailed flow past a periodic array of cylindrical roughness elements. A spectral DNS code was used to model flow over a flat plate surface, and roughness elements were formed using an immersed boundary technique. Solutions were obtained for two roughness heights corresponding to Reynolds numbers (Rek) of 202 and 334, and results are presented for both cases. Cylindrical roughness elements with Rek=202 produced minimal disturbances and the flow appeared generally laminar in the wake downstream of the roughness elements. Flow past cylindrical roughness elements corresponding to Rek=334 was found to transition as soon as 2-3 cylinder diameters downstream and had developed into fully turbulent flow by the end of the domain. Results were found to compare reasonably well with a similar set of DNS computations by Rizzetta and Visbal 1 using a sixth-order-accurate centered compact finite difference scheme as well as experimental results obtained by Ergin and White 2 using timeaveraged hotwire measurements of the velocity components. Nomenclature

Analysis and observation of spacecraft plume/ionosphere interactions during maneuvers of the space shuttle

This work employs in situ measurement data and constructive simulations to examine the underlying physical mechanisms that drive spacecraft plume interactions with the space environment in low-Earth orbit. The study centers on observations of the enhanced flux of plasma generated during a maneuver of Space Shuttle Endeavour as part of the Sensor Test for Orion Relative Navigation Risk Mitigation experiment in May 2011. The Canary electrostatic analyzer (ESA) instrument mounted on the portside truss of the International Space Station indicated an elevated ion current during the shuttle maneuver. The apparent source of enhanced ion current is a result of interaction of the spacecraft thruster plume with the rarefied ambient ionosphere, which generates regions of relatively high density plasma through charge exchange between the neutral plume and ambient ions. To reconstruct this event, unsteady simulation data were generated using a combined direct simulation Monte Carlo/particle-in-cell methodology, which employed detailed charge exchange cross-section data and a magnetic field model. The simulation provides local plasma characteristics at the ESA sensor location, and a sensor model is subsequently used to transform the local properties into a prediction of measured ion current. The predicted and observed total currents are presented as a function of time over a 30 s period of pulsed thruster firings. A strong correlation is observed in the temporal characteristics of the simulated and measured total current, and good agreement is also achieved in the total current predicted by the model. These results support conclusions that (1) the enhanced flux of plasma observed by the ESA instrument is associated with Space Shuttle thruster firings and (2) the simulation model captures the essential features of the plume interactions based on the observation data.

A non-equilibrium surface reservoir approach for hybrid DSMC/Navier-Stokes particle generation

An approach for the generation of particles at a hybrid Navier-Stokes/DSMC interface is presented for simple gases and gas mixtures with internal degrees of freedom. DSMC particles generated at a hybrid boundary are assigned thermal velocities using a non-equilibrium surface reservoir approach, in which the fluxes of mass, momentum and energy determined from the Navier-Stokes solution are used to prescribe the appropriate velocity distribution function used in the DSMC particle generation. The non-equilibrium surface reservoir approach is first outlined for a simple (single-species, monatomic) gas, and is then extended to gas mixtures with internal degrees of freedom, in which additional diffusion and internal heat flux terms are included in the Generalized Chapman-Enskog formulation of the perturbation. The significance of the diffusion, shear stress and heat flux breakdown parameters used to compute the perturbation are examined at a hybrid interface within non-equilibrium boundary layer flow, as well as within the breakdown region near a normal shock, in a five-species air gas mixture. The validity of the Chapman-Enskog perturbation at each of these hybrid interfaces is assessed by comparison with the Generalized Chapman-Enskog perturbations. Although a hybrid flowfield solution is not presented, this work provides a rigorous approach for non-equilibrium particle generation involving general hybrid particle/continuum studies of hypersonic flows.

Consistent treatment of transport properties for five-species air direct simulation Monte Carlo/Navier-Stokes applications

A general approach for achieving consistency in the transport properties between direct simulation Monte Carlo (DSMC) and Navier-Stokes (CFD) solvers is presented for five-species air. Coefficients of species diffusion, viscosity, and thermal conductivities are considered. The transport coefficients that are modeled in CFD solvers are often obtained by expressions involving sets of collision integrals, which are obtained from more realistic intermolecular potentials (i.e., ab initio calculations). In this work, the self-consistent effective binary diffusion and Gupta et al.–Yos tranport models are considered. The DSMC transport coefficients are approximated from Chapman-Enskog theory in which the collision integrals are computed using either the variable hard sphere (VHS) and variable soft sphere (VSS) (phenomenological) collision cross section models. The VHS and VSS parameters are then used to adjust the DSMC transport coefficients in order to achieve a best-fit to the coefficients computed from more re...

Effects of rarefaction on hypersonic boundary layer flow over discrete surface roughness

The effects of rarefaction on hypersonic boundary layer flow over a discrete surface roughness element are examined in the context of the STS-119 flight experiment. The purpose of this flight experiment was to acquire data regarding the onset of roughnessinduced boundary layer transition during atmospheric entry of the Orbiter. Transition onset was confirmed at a freestream Mach number M∞=15. The height of the boundary layer trip k was determined to be approximately one-quarter of the boundary layer thickness δ, and initial estimates indicate that the local Knudsen number (Kn=λ/k) in the region of the roughness was O(10 -3 ), such that k≈450λ, where λ is the molecular mean free path. In this regime, the continuum approximations of zero velocity and no thermal slip at the wall begin to break down, and thermal non-equilibrium effects may become more prominent due to a relative increase in time required for thermal equilibration. The aim of this work is to address the significance of rarefaction effects in modeling the disturbance field generated by hypersonic boundary layer flow over surface roughness using a hybrid of the DAC and DPLR numerical simulation codes. Preliminary studies have been conducted to examine flow over the STS-119 boundary layer trip geometry under flight conditions at M∞=20.3. A comparison of the flowfield quantities between the hybrid and DPLR solutions indicate good agreement in the general shock structure and expansion formed in the region of the roughness, and it was also observed that the disturbances of the flowfield quantities in the wake tend to be more pronounced in the DPLR solution. The total heat flux on the surface of the protuberance is also examined, and preliminary results indicate that the hybrid solution produces a peak heat flux that is approximately 10% lower than the peak heat flux predicted by DPLR.

Development of a hybrid DSMC/Navier-stokes solver with application to the STS-119 boundary layer transition flight experiments

The effects of rarefaction on hypersonic boundary layer flow over a discrete surface roughness element are examined in the context of the STS-119 flight experiment. The purpose of this flight experiment was to acquire data regarding the onset of roughnessinduced boundary layer transition during atmospheric entry of the Orbiter. Transition onset was confirmed at a freestream Mach number M∞ =1 5. 1 Initial estimates indicate that the local Knudsen number (Kn = λ/k) in the region of the roughness was O(10 −3 ) ,s uch that k ∼ 450λ ,w herek is the height of the protuberance and λ is the molecular mean free path. In this regime, the continuum approximations of zero velocity and no thermal slip at the wall begin to break down, and thermal non-equilibrium effects may become more prominent due to a relative increase in time required for thermal equilibration. The aim of this work is to address the significance of rarefaction effects in modeling the disturbance field generated by hypersonic boundary layer flow over surface roughness using a hybrid of the DAC and DPLR numerical simulation codes. A new method for generating DAC particles from a non-equilibrium surface reservoir is presented, and we also outline a method for matching the transport properties between the DAC and DPLR solvers. Studies have been conducted to examine flow over the STS-119 boundary layer trip geometry under flight conditions at M∞ =2 0.3. A comparison of the flowfield quantities between the hybrid and DPLR solutions indicate good agreement in the general shock structure and expansion formed in the region of the roughness, and it was also observed that the disturbances of the flowfield quantities in the wake tend to be more pronounced in the DPLR solution. The heat flux on the surface of the protuberance and the surface downstream of the protuberance is also examined. Both solvers show a significant heating augmentation on the protuberance itself, with a band of high surface heating which originates from the protuberance leading edge and extends downtream. Results from this study indicate that the Navier-Stokes solution predicts a peak heating on the roughness which is 6 times that of the undisturbed surface heating, while the vortex heating in the wake produces an increase in heating which is 3 times greater than the undisturbed surface heating. In contrast, the hybrid solution predicts a peak heating on the protuberance which is 4.5 times that of the undisturbed surface heating, while the vortex heating in the wake increases the surface heating to 1.5 times that of the undisturbed surface heating.

Temperature / rarefaction effects in hypersonic boundary-layer flow with an oblique roughness element

Three-dimensional laminar hypersonic boundary-layer flows are investigated applying the compressible bi-global linear stability theory (B-LST) in flow crossplanes. The flat-plate flow is altered by an obliquely placed discrete fence-like roughness element that is about half the boundary-layer thickness high. Roughness setup and flow conditions resemble the STS-119 flight experiment. A cold-flow case and hot-flow cases are considered. The influence of non-perfect gas properties such as variable chemical composition, or thermal energy relaxation are included. The steady base flows are extracted from Navier-Stokes simulations. The underlying gas modell for reacting and non-reacting air accounts for thermal as well as chemical nonequilibrium. Rarefaction effects are considered in terms of a slip condition for velocity and temperature at the wall. Stability properties of the roughness wake under cold, hot, and hot rarefied flow conditions are compared in terms of local and integral growth.

The effects of trapped bubbles on viscous drag reduction for submerged surfaces

The effect on viscous drag of trapped bubbles on a submerged flat plate was investigated. The objective was to determine if viscous drag reduction could be obtained by replacing portions of the solid no-slip surface of the plate with areas of near-slip formed by bubbles. One configuration involved a large bubble trapped on the bottom surface of a horizontally mounted plate, which provides insight as to the maximum drag reduction obtainable using the trapped bubble concept. A second configuration involved a trapped bubble array (TBA), which used electrolysis to grow and maintain bubbles on the plate surface in thousands of tiny conductive holes. The TBA experiments were conducted on a vertical plate, to demonstrate the versatility of this drag reduction method. Drag measurements were acquired over a range of Reynolds numbers using either a force balance for plates mounted in a vertical orientation, or by performing a momentum integral balance using a LDA wake survey for plates mounted in either vertical or horizontal orientations. Results show that a drag reduction of up to 32% was obtained for the large trapped bubble case, while negligible drag reduction was obtained for the array of tiny trapped bubbles.