Femtosecond laser electronic excitation tagging (FLEET) velocimetry is an important diagnostic technique for seedless velocimetry measurements particularly in supersonic and hypersonic flows. Typical FLEET measurements feature a single laser line and camera system to achieve one-component velocimetry along a line, although some multiple-spot and multiple-component configurations have been demonstrated. In this work, tomographic imaging is used to track the three-dimensional location of many FLEET spots. A quadscope is used to combine four unique views onto a single high-speed image intensifier and camera. Tomographic reconstructions of the FLEET emission are analyzed for three-component velocimetry from multiple FLEET spots. Glass wedges are used to create many (nine) closely spaced FLEET spots with less than 10% transmission losses. These developments lead to a significant improvement in the dimensionality and spatial coverage of a FLEET instrument with some increases in experimental complexity and data processing. Multiple-point three-component FLEET velocimetry is demonstrated in an underexpanded jet.
Fluid–structure interactions were measured between a representative control surface and the hypersonic flow deflected by it. The control surface is simplified as a spanwise finite ramp placed on a longitudinal slice of a cone. The front surface of the ramp contains a thin panel designed to respond to the unsteady fluid loading arising from the shock-wave/boundary-layer interactions. Experiments were conducted at Mach 5 and Mach 8 with ramps of different angles. High-speed schlieren captured the unsteady flow dynamics and accelerometers behind the thin panel measured its structural response. Panel vibrations were dominated by natural modes that were excited by the broadband aerodynamic fluctuations arising in the flowfield. However, increased structural response was observed in two distinct flow regimes: 1) attached or small separation interactions, where the transitional regime induced the strongest panel fluctuations. This was in agreement with the observation of increased convective undulations or bulges in the separation shock generated by the passage of turbulent spots, and 2) large separated interactions, where shear layer flapping in the laminar regime produced strong panel response at the flapping frequency. In addition, panel heating during the experiment caused a downward shift in its natural mode frequencies.
This work presents measurements of liquid drop deformation and breakup time behind approximately conical shock waves and evaluates the predictive capabilities of low-order models and correlations developed using planar shock experiments. A conical shock was approximated by firing a bullet at Mach 4.5 past a vertical column of water drops with a mean initial diameter of 192 µm. The time-resolved drop position and maximum transverse dimension were characterized using backlit stereo images taken at 500 kHz. The gas density and velocity fields experienced by the drops were estimated using a Reynolds-averaged Navier-Stokes simulation of the bullet. Classical correlations predict drop breakup times and deformation in error by a factor of 3 or more. The Taylor analogy breakup (TAB) model predicts deformed drop diameters that agree within the confidence bounds of the ensemble-averaged experimental values using a dimensionless constant C2 = 2 compared to the accepted value C2 = 2/3. Results demonstrate existing correlations are inadequate for predicting the drop response to the three-dimensional relaxation of the flowfield downstream of a conical-like shock and suggest the TAB model results represent a path toward improved predictions.
Measurements of gas-phase temperature and pressure in hypersonic flows are important for understanding gas-phase fluctuations which can drive dynamic loading on model surfaces and to study fundamental compressible flow turbulence. To achieve this capability, femtosecond coherent anti-Stokes Raman scattering (fs CARS) is applied in Sandia National Laboratories’ cold-flow hypersonic wind tunnel facility. Measurements were performed for tunnel freestream temperatures of 42–58 K and pressures of 1.5–2.2 Torr. The CARS measurement volume was translated in the flow direction during a 30-second tunnel run using a single computer-controlled translation stage. After broadband femtosecond laser excitation, the rotational Raman coherence was probed twice, once at an early time where the collisional environment has not affected the Raman coherence, and another at a later time after the collisional environment has led to significant dephasing of the Raman coherent. The gas-phase temperature was obtained primarily from the early-probe CARS spectra, while the gas-phase pressure was obtained primarily from the late-probe CARS spectra. Challenges in implementing fs CARS in this facility such as changes in the nonresonant spectrum at different measurement location are discussed.
This study explores the evolution of a turbulent hypersonic boundary layer over a spanwise-finite expansion-compression geometry. The geometry is based on a slender cone with an axial slice that subjects the cone boundary layer to a favorable pressure gradient. The mean flow field was obtained from a hybrid RANS-LES computation that showed the thickening of the boundary layer, a decrease in the mean pressure and the development of incipient streamwise vortical structures on the slice. The experiments use fluctuating surface pressure and shear-stress sensors along the centerline of the slice which demonstrate significant reduction in turbulence activity on the slice indicating relaminarization of the boundary-layer. These observations were corroborated by high framerate schlieren, filtered Rayleigh scattering and scanning focused laser differential interferometry. When a 10◦ ramp is introduced at the aft end of the slice, the effectively relaminarized boundary-layer separates upstream of the slice-ramp corner due to its increased susceptibility to separation in comparison to a turbulent boundary layer.
This work applies Filtered Rayleigh Scattering (FRS) for the study of shock wave/boundary layer interactions on a cone-slice-ramp geometry. As FRS measures a planar slice of the flow, the three-dimensionality of this geometry can be captured, rather than in path-integrated imaging such as schlieren. A carbon dioxide seeding system designed for the Sandia Hypersonic Wind Tunnel provides sufficient light scattering for FRS measurements. Strong background rejection in the images was achieved using a molecular gas filter, resulting in detailed visualization of flow structures within the boundary and shear layers. Images show separation and reattachment shock, as well as structures related to flow instability and transition to turbulence. A highly unsteady separation region was investigated, showing instantaneous shaping of the shock structure with turbulence.
The Reynolds-averaged Navier–Stokes (RANS) equations remain a workhorse technology for simulating compressible fluid flows of practical interest. Due to model-form errors, however, RANS models can yield erroneous predictions that preclude their use on mission-critical problems. This work presents a data-driven turbulence modeling strategy aimed at improving RANS models for compressible fluid flows. The strategy outlined has three core aspects: (1) prediction for the discrepancy in the Reynolds stress tensor and turbulent heat flux via machine learning (ML), (2) estimating uncertainties in ML model outputs via out-of-distribution detection, and (3) multi-step training strategies to improve feature-response consistency. Results are presented across a range of cases publicly available on NASA’s turbulence modeling resource involving wall-bounded flows, jet flows, and hypersonic boundary layer flows with cold walls. We find that one ML turbulence model is able to provide consistent improvements for numerous quantities-of-interest across all cases.
Previous efforts determined a set of calibrated, optimal model parameter values for Reynolds-averaged Navier–Stokes (RANS) simulations of a compressible jet in crossflow (JIC) using a $k–ε$ turbulence model. These parameters were derived by comparing simulation results to particle image velocimetry (PIV) data of a complementary JIC experiment under a limited set of flow conditions. Here, a $k–ε$ model using both nominal and calibrated parameters is validated against PIV data acquired from a much wider variety of JIC cases, including a realistic flight vehicle. The results from the simulations using the calibrated model parameters showed considerable improvements over those using the nominal values, even for cases that were not used in the calibration procedure that defined the optimal parameters. This improvement is demonstrated using a number of quality metrics that test the spatial alignment of the jet core, the magnitudes of multiple flow variables, and the location and strengths of vortices in the counter-rotating vortex cores on the PIV planes. These results suggest that the calibrated parameters have applicability well outside the specific flow case used in defining them and that with the right model parameters, RANS solutions for the JIC can be improved significantly over those obtained from the nominal model.
This work describes the development and testing of a carbon dioxide seeding system for the Sandia Hypersonic Wind Tunnel. The seeder injects liquid carbon dioxide into the tunnel, which evaporates in the nitrogen supply line and then condenses during the nozzle expansion into a fog of particles that scatter light via Rayleigh scattering. A planar laser scattering (PLS) experiment is conducted in the boundary layer and wake of a cone at Mach 8 to evaluate the success of the seeder. Second-mode waves and turbulence transition were well-visualized by the PLS in the boundary layer and wake. PLS in the wake also captured the expansion wave over the base and wake recompression shock. No carbon dioxide appears to survive and condense in the boundary layer or wake, meaning alternative seeding methods must be explored to extract measurements within these regions. The seeding system offers planar flow visualization opportunities and can enable quantitative velocimetry measurements in the future, including filtered Rayleigh scattering.
Femtosecond laser electronic excitation tagging (FLEET) is a powerful unseeded velocimetry technique typically used to measure one component of velocity along a line, or two or three components from a dot. In this Letter, we demonstrate a dotted-line FLEET technique which combines the dense profile capability of a line with the ability to perform two-component velocimetry with a single camera on a dot. Our set-up uses a single beam path to create multiple simultaneous spots, more than previously achieved in other FLEET spot configurations. We perform dotted-line FLEET measurements downstream of a highly turbulent, supersonic nitrogen free jet. Dotted-line FLEET is created by focusing light transmitted by a periodic mask with rectangular slits of 1.6 × 40 mm2 and an edge-to-edge spacing of 0.5 mm, then focusing the imaged light at the measurement region. Up to seven symmetric dots spaced approximately 0.9 mm apart, with mean full-width at half maximum diameters between 150 and 350 µm, are simultaneously imaged. Both streamwise and radial velocities are computed and presented in this Letter.
This work presents an experimental investigation of the deformation and breakup of water drops behind conical shock waves. A conical shock is generated by firing a bullet at Mach 4.5 past a vertical column of drops with a mean initial diameter of 192 µm. The time-resolved drop position and maximum transverse dimension are characterized using backlit stereo videos taken at 500 kHz. A Reynolds-Averaged Navier Stokes (RANS) simulation of the bullet is used to estimate the gas density and velocity fields experienced by the drops. Classical correlations for breakup times derived from planar-shock/drop interactions are evaluated. Predicted drop breakup times are found to be in error by a factor of three or more, indicating that existing correlations are inadequate for predicting the response to the three-dimensional relaxation of the velocity and thermodynamic properties downstream of the conical shock. Next, the Taylor Analogy Breakup (TAB) model, which solves a transient equation for drop deformation, is evaluated. TAB predictions for drop diameter calculated using a dimensionless constant of C2 = 2, as compared to the accepted value of C2 = 2/3, are found to agree within the confidence bounds of the ensemble averaged experimental values for all drops studied. These results suggest the three-dimensional relaxation effects behind conical shock waves alter the drop response in comparison to a step change across a planar shock, and that future models describing the interaction between a drop and a non-planar shock wave should account for flow field variations.
Measurements of gas-phase pressure and temperature in hypersonic flows are important to understanding fluid–structure interactions on vehicle surfaces, and to develop compressible flow turbulence models. To achieve this measurement capability, femtosecond coherent anti-Stokes Raman scattering (fs CARS) is applied at Sandia National Laboratories’ hypersonic wind tunnel. After excitation of rotational Raman transitions by a broadband femtosecond laser pulse, two probe pulses are used: one at an early time where the collisional environment has largely not affected the Raman coherence, and another at a later time after the collisional environment has led to significant J-dependent dephasing of the Raman coherence. CARS spectra from the early probe are fit for temperature, while the later CARS spectra are fit for pressure. Challenges related to implementing fs CARS in cold-flow hypersonic facilities are discussed. Excessive fs pump energy can lead to flow perturbations. The output of a second-harmonic bandwidth compressor (SHBC) is spectrally filtered using a volume Bragg grating to provide the narrowband ps probe pulses and enable single-shot CARS measurements at 1 kHz. Measurements are demonstrated at temperatures and pressures relevant to cold-flow hypersonic wind tunnels in a low-pressure cryostat with an initial demonstration in the hypersonic wind tunnel.
Time-resolved particle image velocimetry (TR-PIV) has become widespread in fluid dynamics. Essentially a velocity field movie, the dynamic content provides temporal as well as spatial information, in contrast to conventional PIV offering only statistical ensembles of flow quantities. From these time series arise further analyses such as accelerometry, space-time correlations, frequency spectra of turbulence including spatial variability, and derivation of pressure fields and forces. The historical development of TR-PIV is chronicled, culminating in an assessment of the current state of technology in high-repetition-rate lasers and high-speed cameras. Commercialization of pulse-burst lasers has expanded TR-PIV into more flows, including the compressible regime, and has achieved MHz rates. Particle response times and peak locking during image interrogation require attention but generally are not impediments to success. Accuracy considerations are discussed, including the risks of noise and aliasing in spectral content. Oversampled TR-PIV measurements allow use of multi-frame image interrogation methods, which improve the precision of the correlation and raise the velocity dynamic range of PIV. In combination with volumetric methods and data assimilation, a full four-dimensional description of a flow is not only achievable but becoming standardized. A survey of exemplary applications is followed by a few predictions concerning the future of TR-PIV.
The development of new hypersonic flight vehicles is limited by the physical understanding that may be obtained from ground test facilities. This has motivated the present development of a temporally and spatially resolved velocimetry measurement for Sandia National Laboratories (SNL) Hypersonic Wind Tunnel (HWT) using Femtosecond Laser Electronic Excitation Tagging (FLEET). First, a multi-line FLEET technique has been created for the first time and tested in a supersonic jet, allowing simultaneous measurements of velocities along multiple profiles in a flow. Secondly, two different approaches have been demonstrated for generating dotted FLEET lines. One employs a slit mask pattern focused into points to yield a dotted line, allowing for two- or three-component velocity measurements free of contamination between components. The other dotted-line approach is based upon an optical wedge array and yields a grid of points rather than a dotted line. Two successful FLEET measurement campaigns have been conducted in SNL’s HWT. The first effort established optimal diagnostic configurations in the hypersonic environment based on earlier benchtop reproductions, including validation of the use of a 267 nm beam to boost the measurement signal-to-noise ratio (SNR) with minimal risk of perturbing the flow and greater simplicity than a comparable resonant technique at 202 nm. The same FLEET system subsequently was reconstituted to demonstrate the ability to make velocimetry measurements of hypersonic turbulence in a realistic flow field. Mean velocity profiles and turbulence intensity profiles of the shear layer in the wake of a hypersonic cone model were measured at several different downstream stations, proving the viability of FLEET as a hypersonic diagnostic.
The primary parameter of a standard k-ϵ model, Cμ, was calculated from stereoscopic particle image velocimetry (PIV) data for a supersonic jet exhausting into a transonic crossflow. This required the determination of turbulent kinetic energy, turbulent eddy viscosity, and turbulent energy dissipation rate. Image interrogation was optimized, with different procedures used for mean strain rates and Reynolds stresses, to produce useful turbulent eddy viscosity fields. The eddy viscosity was calculated by a least-squares fit to all components of the three-dimensional strain-rate tensor that were available from the PIV data. This eliminated artifacts and noise observed when using a single strain component. Local dissipation rates were determined via Kolmogorov’s similarity hypotheses and the second-order structure function. The eddy viscosity and dissipation rates were then combined to determine Cμ. Considerable spatial variation was observed in Cμ, with the highest values found in regions where turbulent kinetic energy was relatively ow but where turbulent mixing was important, e.g., along the high-strain jet edges and in the wake region. This suggests that use of a constant Cμ in modeling may lead to poor Reynolds stress predictions at mixing interfaces. A data-driven modeling approach that can predict this spatial variation of Cμ based on known state variables may lead to improved simulation results without the need for calibration.
Multi-frame correlation algorithms for time-resolved PIV have been shown in previous studies to reduce noise and error levels in comparison with conventional two-frame correlations. However, none of these prior efforts tested the accuracy of the algorithms in spectral space. Even should a multi-frame algorithm reduce the error of vector computations summed over an entire data set, this does not imply that these improvements are observed at all frequencies. The present study examines the accuracy of velocity spectra in comparison with simultaneous hot-wire data. Results indicate that the high-frequency content of the spectrum is very sensitive to choice of the interrogation algorithm and may not return an accurate response. A top-hat-weighted sliding sum-of-correlation is contaminated by high-frequency ringing whereas Gaussian weighting is indistinguishable from a low-pass filtering effect. Some evidence suggests the pyramid correlation modestly increases bandwidth of the measurement at high frequencies. The apparent benefits of multi-frame interrogation algorithms may be limited in their ability to reveal additional spectral content of the flow.
This experimental study explores the fluid-structure interactions occurring between a control surface and the hypersonic flow deflected by it. The control surface is simplified for this work as a spanwise finite wedge placed on a longitudinally sliced part of the cone. The front surface of the wedge is a thin panel which is designed to respond to the unsteady fluid loading arising from the shock-wave/boundary layer interactions. Experiments have been conducted in the Sandia Hypersonic Wind Tunnel at Mach 5 and Mach 8 at wedge angles of 10◦, 20◦ and 30◦ . High-speed schlieren and backside panel accelerometer measurements capture the unsteady flow dynamics and structural response of the thin panel, respectively. For attached or small separation interactions, the transitional regime has the strongest panel fluctuations with convective shock undulations induced by the boundary layer disturbance shown to be associated with dominant panel vibrations. For large separated interactions, shear layer flapping can excite select panel modes. Heating of the panel causes a downward shift in natural mode frequencies.
This study seeks to simplify the optical requirements for multi-line FLEET (Femtosecond Laser Electronic Excitation Tagging) generation by focusing the image of a periodic slit-mask with a cylindrical and spherical lens. Geometry effects on the signal were analyzed over fifteen mask iterations. The signal for each mask was found to vary with mask standoff from the focusing optics, which was optimized based on maximizing the Signal-to-Noise Ratio (SNR) for each mask. The number of generated lines was found to decrease with slit spacing while the separation of the lines increased. FLEET line spacing was determined by a constant magnification value of the imaged masks’ slit spacing. From the geometry study, two masks that produced three to five lines spaced at 0.8–1 mm apart with SNR > 4 were chosen to demonstrate the multi-line technique in a supersonic free-jet. Velocity calculations from this data showed good agreement with schlieren imaging of compressible flow structures.
A simple linear configuration for multi-line femtosecond laser electronic excitation tagging (FLEET) velocimetry is used for the first time, to the best of our knowledge, to image an overexpanded unsteady supersonic jet. The FLEET lines are spaced 0.5-1.0mmapart, and up to six lines can be used simultaneously to visualize the flowfield. These lines are created using periodic masks, despite the mask blocking 25%-30%of the 10 mJ incident beam.Maps of mean singlecomponent velocity in the direction along the principal flow axis, and turbulence intensity in that same direction, are created using multi-line FLEET, and computed velocities agree well with those obtained from single-line (traditional) FLEET. Compared to traditional FLEET, multi-line FLEET offers increased simultaneous spatial coverage and the ability to produce spatial correlations in the streamwise direction. This FLEET permutation is especially well suited for short-duration test facilities.
An experimental characterization of the flow environment for the Sandia Axisymmetric Transonic Hump is presented. This is an axisymmetric model with a circular hump tested at a transonic Mach number, similar to the classic Bachalo-Johnson configuration. The flow is turbulent approaching the hump and becomes locally supersonic at the apex. This leads to a shock-wave/boundary-layer interaction, an unsteady separation bubble, and flow reattachment downstream. The characterization focuses on the quantities required to set proper boundary conditions for computational efforts described in the companion paper, including: 1) stagnation and test section pressure and temperature; 2) turbulence intensity; and 3) tunnel wall boundary layer profiles. Model characterization upstream of the hump includes: 1) surface shear stress; and 2) boundary layer profiles. Note: Numerical values characterizing the experiment have been redacted from this version of the paper. Model geometry and boundary conditions will be withheld until the official start of the Validation Challenge, at which time a revised version of this paper will become available. Data surrounding the hump are considered final results and will be withheld until completion of the Validation Challenge.
Bench-top tests are conducted to characterize Femtosecond Laser Electronic Excitation Tagging (FLEET) in static low pressure (35 mTorr-760 Torr) conditions, and to measure the acoustic disturbance caused by the resulting filament as a function of tagging wavelength and energy. The FLEET line thickness as a function of pressure and delay is described by a simple diffusion model. Initial FLEET measurements in a Mach 8 flow show that gate times of ≥ 1µs can produce visible smearing of the FLEET emission and challenge the traditional Gaussian fitting methods used to find the line center. To minimize flow perturbations and uncertainty of the final line position, several recommendations are offered: using third harmonic FLEET at 267 nm for superior signal levels with lower energy deposition than both 800 nm and 400 nm FLEET, and short camera delays and exposure times to reduce fitting uncertainty. This guidance is implemented in a Mach 8 test condition and results are presented.
Two techniques have extended the effective frequency limits of postage-stamp PIV, in which a pulse-burst laser and very small fields of view combine to achieve high repetition rates. An interpolation scheme reduced measurement noise, raising the effective frequency response of previous 400-kHz measurements from about 120 kHz to 200 kHz. The other technique increased the PIV acquisition rate to very nearly MHz rates (990 kHz) by using a faster camera. Charge leaked through the camera shift register at these framing rates but this was shown not to bias the measurements. The increased framing rate provided oversampled data and enabled use of multi-frame correlation algorithms for a lower noise floor, increasing the effective frequency response to 240 kHz where the interrogation window size begins to spatially filter the data. Good agreement between the interpolation technique and the MHz-rate PIV measurements was established. The velocity spectra suggest turbulence power-law scaling in the inertial subrange steeper than the theoretical-5/3 scaling, attributed to an absence of isotropy.
Here, experiments were performed within Sandia National Labs’ Multiphase Shock Tube to measure and quantify the shock-induced dispersal of a shock/dense particle curtain interaction. Following interaction with a planar travelling shock wave, schlieren imaging at 75 kHz was used to track the upstream and downstream edges of the curtain. Data were obtained for two particle diameter ranges ($d_{p}=106{-}125$,$300{-}355~\unicode[STIX]{x03BC}\text{m}$) across Mach numbers ranging from 1.24 to 2.02. Using these data, along with data compiled from the literature, the dispersion of a dense curtain was studied for multiple Mach numbers (1.2–2.6), particle sizes ($100{-}1000~\unicode[STIX]{x03BC}\text{m}$) and volume fractions (9–32 %). Data were non-dimensionalized according to two different scaling methods found within the literature, with time scales defined based on either particle propagation time or pressure ratio across a reflected shock. The data refelct that spreading of the particle curtain is a function of the volume fraction, with the effectiveness of each time scale based on the proximity of a given curtain’s volume fraction to the dilute mixture regime. It is observed that volume fraction corrections applied to a traditional particle propagation time scale result in the best collapse of the data between the two time scales tested here. In addition, a constant-thickness regime has been identified, which has not been noted within previous literature.
Time-resolved particle image velocimetry was conducted at 40 kHz using a pulse-burst laser in the supersonic wake of a wall-mounted hemisphere. Velocity fields suggest a recirculation region with two lobes, in which flow moves away from the wall near the centerline and recirculates back toward the hemisphere off the centerline, contrary to transonic configurations. Spatio-temporal cross-correlations and conditional ensemble averages relate the characteristic behavior of the unsteady shock motion to the flapping of the shear layer. At Mach 1.5, oblique shocks develop, associated with vortical structures in the shear layer and convect downstream in tandem; a weak periodicity is observed. Shock motion at Mach 2.0 appears somewhat different, wherein multiple weak disturbances propagate from shear-layer turbulent structures to form an oblique shock that ripples as these vortices pass by. Bifurcated shock feet coalesce and break apart without evident periodicity. Power spectra show a preferred frequency of shear-layer flapping and shock motion for Mach 1.5, but at Mach 2.0, a weak preferred frequency at the same Strouhal number of 0.32 is found only for oblique shock motion and not shear-layer unsteadiness.
Femtosecond Laser Electronic Excitation Tagging (FLEET) is used to measure velocity flowfields in the wake of a sharp 7◦ half-angle cone in nitrogen at Mach 8, over freestream Reynolds numbers from 4.3∗106 /m to 13.8∗106 /m. Flow tagging reveals expected wake features such as the separation shear layer and two-dimensional velocity components. Frequency-tripled FLEET has a longer lifetime and is more energy efficient by tenfold compared to 800 nm FLEET. Additionally, FLEET lines written with 267 nm are three times longer and 25% thinner than that written with 800 nm at a 1 µs delay. Two gated detection systems are compared. While the PIMAX 3 ICCD offers variable gating and fewer imaging artifacts than a LaVision IRO coupled to a Photron SA-Z, its slow readout speed renders it ineffective for capturing hypersonic velocity fluctuations. FLEET can be detected to 25 µs following excitation within 10 mm downstream of the model base, but delays greater than 4 µs have deteriorated signal-to-noise and line fit uncertainties greater than 10%. In a hypersonic nitrogen flow, exposures of just several hundred nanoseconds are long enough to produce saturated signals and/or increase the line thickness, thereby adding to measurement uncertainty. Velocity calculated between the first two delays offer the lowest uncertainty (less than 3% of the mean velocity).
Previous efforts determined a set of calibrated model parameters for ReynoldsAveraged Navier Stokes (RANS) simulations of a compressible jet in crossflow (JIC) using a k-ɛ turbulence model. These coefficients were derived from Particle Image Velocimetry (PIV) data of a complementary experiment using a limited set of flow conditions. Here, k-ɛ models using conventional (nominal) and calibrated parameters are rigorously validated against PIV data acquired under a much wider variety of JIC cases, including a flight configuration. The results from the simulations using the calibrated model parameters showed considerable improvements over those using the nominal values, even for cases that were not used in defining the calibrated parameters. This improvement is demonstrated using quality metrics defined specifically to test the spatial alignment of the jet core as well as the magnitudes of flow variables on the PIV planes. These results suggest that the calibrated parameters have applicability well outside the specific flow case used in defining them and that with the right model parameters, RANS results can be improved significantly over the nominal.
Fluid-structure interactions were studies on a 7° half-angle cone in the Sandia Hypersonic Wind Tunnel at Mach 5 and 8 and in the Purdue Boeing/AFOSR Mach 6 Quiet Tunnel. A thin composite panel was integrated into the cone and the response to boundary-layer disturbances was characterized by accelerometers on the backside of the panel. Here, under quiet-flow conditions at Mach 6, the cone boundary layer remained laminar. Artificially generated turbulent spots excited a directionally dependent panel response which would last much longer than the spot duration.
The mechanism by which aerodynamic effects of jet/fin interaction arise from the flow structure of a jet in crossflow is explored using particle image velocimetry measurements of the crossplane velocity field as it impinges on a downstream fin instrumented with high-frequency pressure sensors. A Mach 3.7 jet issues into a Mach 0.8 crossflow from either a normal or inclined nozzle, and three lateral fin locations are tested. Conditional ensemble-averaged velocity fields are generated based upon the simultaneous pressure condition. Additional analysis relates instantaneous velocity vectors to pressure fluctuations. The pressure differential across the fin is driven by variations in the spanwise velocity component, which substitutes for the induced angle of attack on the fin. Pressure changes at the fin tip are strongly related to fluctuations in the streamwise velocity deficit, wherein lower pressure is associated with higher velocity and vice versa. The normal nozzle produces a counter-rotating vortex pair that passes above the fin, and pressure fluctuations are principally driven by the wall horseshoe vortex and the jet wake deficit. The inclined nozzle produces a vortex pair that impinges the fin and yields stronger pressure fluctuations driven more directly by turbulence originating from the jet mixing.
Turbulent viscosities have been calculated from stereoscopic particle image velocimetry (PIV) data for a supersonic jet exhausting into a transonic crossflow. Image interrogation must be optimized to produce useful turbulent viscosity fields. High-accuracy image reconstruction should be used for the final iteration, whereas efficient algorithms produce spatial artifacts in derivative fields. Mean strain rates should be calculated from large windows (128 pixel) with 75% overlap. Turbulent stresses are optimally computed using multiple (more than two) iterations of image interrogation and 75% overlap, both of which increase the signal bandwidth. However, the improvement is modest and may not justify the considerable increase in computational expense. The turbulent viscosity may be expressed in tensor notation to include all three axes of velocity data. In this formulation, a least-squares fit to the multiple equations comprising the tensor generated a scalar turbulent viscosity that eliminated many of the artifacts and noise present in the single-component formulation. The resulting experimental turbulent viscosity fields will be used to develop data-driven turbulence models that can improve the fidelity of predictive computations.
A new approach to denoising Time-Resolved Particle Image Velocimetry data is proposed by incorporating measurement uncertainties estimated using the correlation statistics method. The denoising algorithm of Oxlade et al (Experiments in Fluids, 2012) has been modified to add the frequency dependence of PIV noise by obtaining it from the uncertainty estimates, including the correlated term between velocity and uncertainty that is zero only if white noise is assumed. Although the present approach was only partially effective in denoising the 400-kHz “postage-stamp PIV” data, important and novel insights were obtained into the behavior of PIV uncertainty. The belief that PIV noise is white noise has been shown to be inaccurate, though it may serve as a reasonable approximation for measurements with a high dynamic range. Noise spectra take a similar shape to the velocity spectra because increased velocity fluctuations correspond to higher shear and therefore increased uncertainty. Coherence functions show that correlation between velocity fluctuations and uncertainty is strongest at low and mid frequencies, tapering to a much weaker correlation at high frequencies where turbulent scales are small with lower shear magnitudes.
The spanwise variation of resonance dynamics in the Mach 0.94 flow over a finite-span cavity of variable length-to-width ratio was explored using time-resolved particle image velocimetry (TR-PIV) in a planform plane above the cavity and time-resolved pressure sensitive paint (TR-PSP) on the floor and adjacent exterior surface. The TR-PIV showed a significant variation in resonant fluctuations to occur across the span of the cavity, which appears to arise from spillage vortices stemming from finite width effects. Thus, the spanwise variation was a strong function of the cavity aspect ratio and was only weakly dependent on the cavity mode number. Modal streamwise velocity fluctuations in the spillage vortices showed large peaks at modes one through three, indicating that resonance dynamics, and not just broadband turbulence effects, are prevalent near the sidewalls. Large peaks in modal pressures were also present on the walls just outside of the cavity. Interestingly, prominent peaks at the mode frequencies were observed in the spanwise velocity spectra as well. These peaks were strongest near the cavity sidewalls suggesting a coupling between the resonance mechanism and the spillage vortices.
Fluid-structure interactions were studied on a 7◦ half-angle cone in the Sandia Hypersonic Wind Tunnel at Mach 5 and 8 and in the Purdue Boeing/AFOSR Mach 6 Quiet Tunnel. A thin composite panel was integrated into the cone and the response to boundary-layer disturbances was characterized by accelerometers on the backside of the panel. Under quiet-flow conditions at Mach 6, the cone boundary layer remained laminar. Artificially generated turbulent spots excited a directionally dependent panel response which would last much longer than the spot duration. When the spot generation frequency matched a structural natural frequency of the panel, resonance would occur and responses over 200 times greater than under a laminar boundary layer were obtained. At Mach 5 and 8 under noisy flow conditions, natural transition driven by the wind-tunnel acoustic noise dominated the panel response. An elevated vibrational response was observed during transition at frequencies corresponding to the distribution of turbulent spots in the transitional flow. Once turbulent flow developed, the structural response dropped because the intermittent forcing from the spots no longer drove panel vibration.
High-speed, time-resolved particle image velocimetry with a pulse-burst laser was used to measure the gas-phase velocity upstream and downstream of a shock wave-particle curtain interaction at three shock Mach numbers (1.22, 1.40, and 1.45) at a repetition rate of 37.5 kHz. The particle curtain was formed from free-falling soda-lime particles resulting in volume fractions of 9% or 23% at mid-height, depending on particle diameter (106-125 and 300-355 μm, respectively). Following impingement by a shock wave, a pressure difference was created between the upstream and downstream sides of the curtain, which accelerated flow through the curtain. Jetting of flow through the curtain was observed downstream once deformation of the curtain began, demonstrating a long-term unsteady effect. Using a control volume approach, the unsteady drag on the curtain was estimated from velocity and pressure data. The drag imposed on the curtain has a strong volume fraction dependence with a prolonged unsteadiness following initial shock impingement. In addition, the data suggest that the resulting pressure difference following the propagation of the reflected and transmitted shock waves is the primary component to curtain drag.
Pulse-burst particle image velocimetry has been used to acquire time-resolved data at 37.5 kHz of the flow over a finite-width rectangular cavity at Mach 0.8. Power spectra of the particle image velocimetry data reveal four resonance modes that match the frequencies detected simultaneously using high-frequency wall pressure sensors, but whose magnitudes exhibit spatial dependence throughout the cavity. Spatiotemporal cross correlations of velocity to pressure were calculated after bandpass filtering for specific resonance frequencies. Cross-correlation magnitudes express the distribution of resonance energy, revealing local maxima and minima at the edges of the shear layer attributable to wave interference between downstream-and upstream-propagating disturbances. Turbulence intensities were calculated using a triple decomposition and are greatest in the core of the shear layer for higher modes, where resonant energies ordinarily are lower. Most of the energy for the lowest mode lies in the recirculation region and results principally from turbulence rather than resonance. Together, the velocity-pressure cross correlations and the triple-decomposition turbulence intensities explain the sources of energy identified in the spatial distributions of power spectra amplitudes.
Time-resolved particle image velocimetry (PIV) was conducted at 40 kHz using a pulse-burst laser in the supersonic wake of a wall-mounted hemisphere. Velocity fields suggest a recirculation region with two lobes in which flow moves away from the wall near centerline and recirculates back towards the hemisphere off centerline. Spatio-temporal cross-correlations and conditional ensemble averages relate the characteristic behavior of the unsteady shock motion to the flapping of the shear layer. At Mach 1.5, oblique shocks form associated with vortical structures in the shear layer and convect downstream in tandem; a weak periodicity is observed. Shock motion at Mach 2.0 appears somewhat different, wherein multiple weak disturbances propagate from shear layer turbulent structures to form an oblique shock that ripples as these vortices pass by. Bifurcated shock feet coalesce and break apart without evident periodicity. Power spectra show a preferred frequency of shear layer flapping and shock motion for Mach 1.5, but at Mach 2.0 a weak preferred frequency is found only for the oblique shock motion and not the shear layer unsteadiness.
Fluid-structure interactions were studied on a store with tunable structural natural frequencies in complex cavity flow. Different leading edge geometries, doors, and internal inserts were used to generate cavity pressure fields that were more representative of an actual aircraft bay. The store loading and response was characterized using point pressure and accelerometer measurements. These data were supplemented with high-frequency pressure-sensitive paint applied to both the store and to the cavity floor to capture the three-dimensional nature of the pressure field in the complex configurations. The natural frequencies of the store were then changed to allow a systematic study of mode matching between the structural natural frequencies and the dominant cavity tone frequencies. In the complex cavities, the store responded to the cavity resonant tones not only in the streamwise and wall-normal directions, but also the spanwise direction. That spanwise response to cavity tones was not observed for previous studies in a simple rectangular cavity, because the flow across the store width in the spanwise direction was uniform. This different behavior highlights the importance of using a representative bay geometry for prediction of the structural response of a store in a flight environment.
The resonance modes in Mach 0.94 turbulent flow over a cavity having a length-to-depth ratio of five were explored using time-resolved particle image velocimetry and time-resolved pressure sensitive paint. Mode-switching occurred in the velocity field simultaneous with the pressure field. The first cavity mode corresponded to large-scale motions in shear layer and in the vicinity of the recirculation region, whereas the second and third modes contained organized structures associated with shear layer vortices. Modal surface pressures exhibited streamwise periodicity generated by the interference of downstream-traveling disturbances in shear layer with upstream-traveling acoustical waves. Because of this interference, the modal velocity fields also exhibited local maxima at locations containing pressure minima and vice-versa. Modal convective (phase) velocities, based on cross-correlations of bandpass-filtered velocity fields, decreased with decreasing mode number as the modal activity resided in lower portions of the cavity. These phase velocities also exhibited streamwise periodicity caused by wave interference. The measurements demonstrate that despite the complexities inherent in compressible cavity flows, many of the most prevalent resonance dynamics can be described with simple acoustical analogies.
Time-resolved particle image velocimetry recently has been demonstrated in high-speed flows using a pulse-burst laser at repetition rates reaching 50 kHz. Turbulent behavior can be measured at still higher frequencies if the field of view is greatly reduced and lower laser pulse energy is accepted. Current technology allows image acquisition at 400 kHz for sequences exceeding 4,000 frames, but for an array of only 128 × 120 pixels, giving the moniker of “postage-stamp PIV.” The technique has been tested far downstream of a supersonic jet exhausting into a transonic crossflow. Two-component measurements appear valid until 100 kHz at which point a noise floor emerges dependent upon the reduction of peak locking. Stereoscopic measurement offers three-component data for turbulent kinetic energy spectra, but exhibits a reduced signal bandwidth and higher noise in the out-of-plane component due to the oblique camera images. The resulting spectra reveal two regions exhibiting power-law dependence describing the turbulent decay. One is the well-known inertial subrange with a slope of -5/3 at high frequencies. The other displays a -1 power-law dependence for a decade of mid-range frequencies corresponding to the energetic eddies measured by PIV, which appears to have been previously unrecognized for high-speed free shear flows.
Time-resolved particle image velocimetry (PIV) was conducted at 40 kHz using a pulse-burst laser in the supersonic wake of a wall-mounted hemisphere. Velocity fields suggest a recirculation region with two lobes in which flow moves away from the wall near centerline and recirculates back towards the hemisphere off centerline. Spatio-temporal cross-correlations and conditional ensemble averages relate the characteristic behavior of the unsteady shock motion to the flapping of the shear layer. At Mach 1.5, oblique shocks form associated with vortical structures in the shear layer and convect downstream in tandem; a weak periodicity is observed. Shock motion at Mach 2.0 appears somewhat different, wherein multiple weak disturbances propagate from shear layer turbulent structures to form an oblique shock that ripples as these vortices pass by. Bifurcated shock feet coalesce and break apart without evident periodicity. Power spectra show a preferred frequency of shear layer flapping and shock motion for Mach 1.5, but at Mach 2.0 a weak preferred frequency is found only for the oblique shock motion and not the shear layer unsteadiness.
The development of the unsteady pressure field on the floor of a rectangular cavity was studied at Mach 0.9 using high-frequency pressure-sensitive paint. Power spectral amplitudes at each cavity resonance exhibit a spatial distribution with an oscillatory pattern; additional maxima and minima appear as the mode number is increased. This spatial distribution also appears in the propagation velocity of modal pressure disturbances. This behavior was tied to the superposition of a downstream-propagating shear-layer disturbance and an upstream-propagating acoustic wave of different amplitudes and convection velocities, consistent with the classical Rossiter model. The summation of these waves generates an interference pattern in the spatial pressure amplitudes and resulting phase velocity of the resonant pressure fluctuations.
Time-resolved particle image velocimetry recently has been demonstrated in high-speed flows using a pulse-burst laser at repetition rates reaching 50 kHz. Turbulent behavior can be measured at still higher frequencies if the field of view is greatly reduced and lower laser pulse energy is accepted. Current technology allows image acquisition at 400 kHz for sequences exceeding 4,000 frames, but for an array of only 128 × 120 pixels, giving the moniker of “postage-stamp PIV.” The technique has been tested far downstream of a supersonic jet exhausting into a transonic crossflow. Two-component measurements appear valid until 100 kHz at which point a noise floor emerges dependent upon the reduction of peak locking. Stereoscopic measurement offers three-component data for turbulent kinetic energy spectra, but exhibits a reduced signal bandwidth and higher noise in the out-of-plane component due to the oblique camera images. The resulting spectra reveal two regions exhibiting power-law dependence describing the turbulent decay. One is the well-known inertial subrange with a slope of -5/3 at high frequencies. The other displays a -1 power-law dependence for a decade of mid-range frequencies corresponding to the energetic eddies measured by PIV, which appears to have been previously unrecognized for high-speed free shear flows.
Three stereoscopic PIV experiments have been examined to test the effectiveness of self-calibration under varied circumstances. Measurements taken in a streamwise plane yielded a robust self-calibration that returned common results regardless of the specific calibration procedure, but measurements in the crossplane exhibited substantial velocity bias errors whose nature was sensitive to the particulars of the self-calibration approach. Self-calibration is complicated by thick laser sheets and large stereoscopic camera angles and further exacerbated by small particle image diameters and high particle seeding density. Despite the different answers obtained by varied self-calibrations, each implementation locked onto an apparently valid solution with small residual disparity and converged adjustment of the calibration plane. Therefore, the convergence of self-calibration on a solution with small disparity is not sufficient to indicate negligible velocity error due to the stereo calibration.