Shock-induced Flow Separation Research Articles

Design considerations for efficient spanwise-inclined air-jet vortex generators for separation control in supersonic and hypersonic flows

Shock-wave/boundary-layer interactions and associated flow separation are frequently occurring phenomena in many high-speed flow applications. Injection of spanwise-inclined air-jets is an efficient means to control shock-induced flow separation in supersonic and transonic flows. Separation-control studies in hypersonic flow are scarce and so far mostly restricted to microramp control. The objectives of the current work are to broaden the scope into hypersonic regime, investigate the influence of crossflow Mach and Reynolds number on the jet-injection flow field, and to find a suitable control parameter that is valid across a wide range of flow conditions. For this purpose, we study injection of a single spanwise-inclined jet in supersonic and hypersonic crossflows using large-eddy simulations. We analyzed the flow topology, induced vortical structures and boundary-layer statistics. The general characteristics of the jet/crossflow interactions are similar in the supersonic and hyperonic regimes: the injection of a spanwise-inclined jet induces an asymmetric flow topology with a complex system of shocks and vortical structures. We identified the ratio of injection-pressure-to-freestream-pressure as a suitable parameter to characterize and compare jet/crossflow interactions in different flow regimes. In addition, we recommend larger boundary-layer-to-jet-diameter ratio for control applications, because it delays the lift-off of major vortices. The present results suggest that rows of spanwise-inclined jets can be efficiently used to control separation also in hypersonic flow. Moreover, low momentum-flux jet injection can be potentially also used for cooling purposes.

Mitigation of Shock-Induced Separation Using Square-Shaped Micro-Serrations—A Preliminary Study

Suppressing shock-induced flow separation has been a long-standing problem in the design of supersonic vehicles. To reduce the structural and design complexity of control devices, a passive control technique based on micro-serrations is proposed and its controlling effects are preliminarily investigated under test conditions in which the Mach number is 2.5 and the ramp creating an incident shock is 15 deg. Meanwhile, a vorticity-based criterion for assessing separation scales is developed to resolve the inapplicability of the zero skin friction criterion caused by wall unevenness. The simulations demonstrate that the height of the first stair significantly influences the separation length. Generally, the separation length is shorter at higher stairs, but when the height is greater than half of the thickness of the incoming boundary layer, the corresponding separation point moves upstream. A stair with a height of only 0.4 times the thickness of the boundary layer reduces the separation length by 2.69%. Further parametric analysis reveals that while the remaining serrations have limited effects on the flow separation, an optimization of their shape (depth and width) can create more favorable spanwise vortices and offer a modest improvement of the overall controlling performance. Compared to the plate case, a 9.13% reduction in the separation length can be achieved using a slightly serrated design in which the leading stair is 0.1 high and the subsequent serrations are 0.2 deep and 0.05 wide (nondimensionalized, with the thickness of the incoming boundary layer). Meanwhile, the micro-serration structure even brings less drag. Considering the minor modification to the structure, the proposed method has the potential for use in conjunction with other techniques to exert enhanced control on separations.

Forced Injection of a Spanwise-Inclined Jet in Supersonic Crossflow

Beneficial effects of forced air-jet vortex generators in controlling shock-induced flow separation motivated us to study forced jet injection in detail to understand the advantageous control effects. We performed implicit large-eddy simulations to examine the influence of sinusoidally forced injection of a spanwise-inclined jet in supersonic crossflow. Two forcing frequencies were considered (with and 0.0062) in a crossflow at Mach 2.5 and momentum-thickness Reynolds number of 7000. The underexpansion of the jet has a near-linear dependence on the injection pressure. Consequently, in the near field, the major vortices pulsate with the injection pressure. Statistical analysis using triple decomposition revealed that periodic oscillations of the major counter-rotating vortex pair (CVP) result in momentum transfer by major CVPs, also on their lateral sides, leading to enhanced mixing and lateral spreading, which is beneficial for separation control. Three-dimensional dynamic mode decomposition revealed that forced injection induces a large-scale clockwise rotation of the major vortex. These effects are weak at low forcing frequency. High-frequency jet forcing thus leads to more efficient control of flow separation. On the other hand, periodic oscillations cause these vortical structures to dissipate and breakdown faster in the midfield region than for steady jets. The placement of forced jets is therefore an important parameter when used for separation-control purposes.

Mitigation of shock-induced flow separation over an axisymmetric flared body using ramped vanes

Mitigation of shock-induced flow separation over an axisymmetric flared body using ramped vanes

Validation of numerical simulations for investigating flow control in hypersonic inlets

Various flow control devices have been conceived for hypersonic inlets to suppress the inlet unstart and reduce the total pressure loss over the wide range of operating conditions of hypersonic systems. In particular, bleed systems that remove the low-momentum portion of the flow have drawn attention because they effectively reduce shock-induced flow separation and suppress inlet unstart. Numerical simulations are crucial for investigating the effectiveness of different bleed configurations in hypersonic inlets during the design phase. However, the flow structure in these inlets and the effect of the bleed on the flow are challenging to predict due to complicated factors such as shock-wave/boundary-layer interactions and compressibility effects. Moreover, since the boundary conditions, turbulence model, and turbulence corrections affect the numerical results, the most appropriate model can be difficult to determine. Therefore, a numerical solver was developed and evaluated on a series of validation cases with available experimental data to establish the fidelity of the numerical simulations. Seven validation cases, ranging from an elementary configuration involving a flat plate to a complex configuration involving a flat plate with a porous bleed and plenum chamber, were investigated. The flat plate flow numerical results, including the turbulent boundary layer and shock-wave/laminar boundary-layer interactions, were consistent with previous experimental and numerical results. However, in the hypersonic inlet cases, the actual physical phenomena could not be represented without considering compressibility and streamline curvature effects, which depend on the hypersonic inlet configuration. Thus, the effects of turbulence corrections on the numerical results of various hypersonic inlet configurations were examined. For cases with rapid expansion near the inlet throat or inlets with curved walls, compressibility or rotation/curvature corrections were necessary to accurately capture the size of the recirculation zone and reproduce the wall pressure data and shock structure along the flow path. However, for bleed flows on flat plates, the simulations were accurate without applying turbulence corrections. Thus, these turbulence corrections should be applied only when rapid expansion occurs or the wall near the shock-induced separation zone is curved. This paper provides the best practice appropriate for various supersonic, hypersonic flow cases, and can be utilized in future numerical studies on hypersonic inlet boundary layer control.

Large-Amplitude Gusts on the NASA Common Research Model

To date, there is limited knowledge on the topic of unsteady nonlinear transonic flow caused by large-amplitude excitations. The Certification Specifications for Large Aeroplanes, however, demand loads computations for large-amplitude excitations, e.g., for gust encounters of an aircraft. In this paper, numerical solutions to the unsteady Reynolds-averaged Navier–Stokes equations will be used to analyze gust responses of a transport aircraft configuration for a variety of sinusoidal gusts with different lengths and amplitudes at different Mach numbers. On the one hand, it is shown that unsteady nonlinear responses can lead to lower maximum lift values compared to their time-linearized reference due to unsteady shock-induced flow separation. This nonlinear effect dominates for large gust lengths and higher Mach numbers. On the other hand, for the lowest Mach number considered, the coalescence of a double shock at medium gust lengths introduces nonlinear effects leading to an overprediction of the time-linearized maximum lift by the nonlinear computations. Though the turbulence model of Spalart and Allmaras shows completely different separation patterns than a differential Reynolds stress model, the variation of the turbulence model does not alter the overall qualitative findings, but has an effect on the quantitative values.

Spanwise-Inclined Jets in Supersonic Crossflow: Effects of Injection Pressure and Separation-Control Effectiveness

Motivated by the advantageous performance of air-jet vortex generators (AJVGs) in controlling shock-induced flow separation, we aim at understanding the effects of relevant injection parameters on the physical mechanisms involved, as well as on the control effectiveness. We performed implicit large-eddy simulations to examine the influence of injection pressure on a spanwise-inclined jet in supersonic crossflow at Mach 2.5 and with momentum-thickness Reynolds number 7000. Four injection pressures were considered. The jet and jet/crossflow interactions strengthen with injection pressure, which is reflected in an enhanced obliqueness and asymmetry of the flow topology, strengthening the induced major counter-rotating vortex pair (CVP). This strengthening of the major CVP stagnates at a certain threshold, above which the flow blockage and momentum influx caused by the jets become so strong that they counter the positive influence on the CVP. Subsequently, we installed these jets in an AJVG control setup for a 24 deg compression-ramp interaction with separation to analyze the control effectiveness and interplay of the injection pressure and jet spacing. The strongest effects on control effectiveness are not related to the injection pressure itself, but to interactions between the jet-induced shocks and vortical structures. Therefore, there is a strong link between jet spacing and injection pressure regarding the control effectiveness. Based on these observations, we provide recommendations for designing efficient AJVG control setups.

Flow control effect of pulsed arc discharge plasma actuation on impinging shock wave/boundary layer interaction

The current study investigates the control effect of the pulsed arc discharge plasma on the impinging shock wave and boundary layer interaction (SWBLI) generated by a 14° wedge in a Mach 2.5 flow. The response characteristics of SWBLI on pulsed arc discharge actuation were illustrated, and the controlling mechanism of shock-induced flow separation under different plasma power settings was revealed. The results, which were well validated by the relative published experiment, showed that when setting the exciting power density ph as 1.0 × 1011 W/m3, the oblique shock wave obtained an obvious fluctuation, and the foot of the reattachment shock wave was partially removed. In addition, as the controlling gas bubble passed through the interaction region, the reverse flow zone was enlarged, and the separation shock wave was shifted upward. When ph was set to 4.8 × 109 W/m3, the flow separation induced by SWBLI was effectively suppressed and the size of the reverse flow zone was significantly reduced. Moreover, as the energy input was increased, the arc-induced blast wave (BW) velocity was obviously enhanced. Additionally, it is further found that the arc plasma energy deposition density in the discharge region was the determining factor for SWBLI control, even for a relatively small exciting energy input. Better drag reduction of the flow field would be achieved with the pulsed arc discharge plasma of higher power density, and a drag reduction rate of nearly 10.05% was obtained at ph = 1 × 1011 W/m3 control condition.

Effect of Flap Deflection on Single-Expansion-Ramp Nozzles Performance at Different Pressure Ratios

Experiments have been carried out to investigate the various performance characteristics of single-expansion-ramp nozzle with straight and deflected flap conditions for two different flap lengths. Experiments were performed at different nozzle pressure ratios (NPRs) ranging between 1.5 and 9. Wall static pressure measurements were made to estimate the various performance parameters, such as surface axial force, normal force, overall axial force, pitching moment, and thrust vectoring angles. Schlieren flow visualization technique was adopted to understand the internal shock-induced flow separation and shock patterns inside the nozzle under various NPRs. Flap deflection and NPR control the shock location through area variation, and thus the pressure distribution on the walls. At low NPRs, the internal drag force generated by deflected flap is lesser than that by straight flap, whereas at high NPRs the straight flap produces significantly higher surface axial thrust over the deflected flap. Behavior of normal force and pitching moment was affected by two factors: flap length and shock location. The straight flap shows significantly higher normal force than deflected flap. Hence, the thrust vectoring was better for straight flap nozzles compared to deflected flap nozzles, over the full range of NPRs tested. On the other hand, the gross thrust force performance is better for the straight flap for low NPRs and for the deflected flap for high NPRs. Overall, at high speeds, longer deflected flap appears to give better performance considering all the parameters, but will lead to heavier nozzle.

Modal Analysis of Separation Bubble Unsteadiness in Conical Shock Wave/Turbulent Boundary Layer Interaction

The frequency properties and the typical flow features within the interaction zone in conical shock-wave/turbulent boundary-layer interactions (CSBLIs) are analyzed by performing proper orthogonal decomposition (POD) and dynamic mode decomposition (DMD) on a time-resolved direct numerical simulation database. The streamwise velocity and pressure fluctuations are processed according to their energy and frequency, respectively. As is the case of planar shock-wave/boundary-layer interactions, fluctuations in some bands of frequencies are more energetic downstream of the flow separation zone. The energetic low-frequency modes of the streamwise velocity fluctuations are reminiscent of Görtler-like vortices, whereas those of pressure fluctuations are connected with the “breathing” motion of the separation bubble. Higher-frequency modes, on the other hand, are related to wrinkling of the impinging conical shock. Although they bear resemblance with those in planar shock/turbulent boundary-layer interactions, it was found that the flow structures herein are less coherent owing to spanwise inhomogeneity of the impinging shock strength. The POD and DMD results of CSBLI support the notion that Görtler-like vortices might be an intrinsic mechanism of the interaction zone, which is probably responsible for the low-frequency unsteadiness of CSBLI. The present results may serve as a basic guidance for the design of fluidic control strategies of shock-induced flow separation.

Control of flow separation over an axisymmetric flared body using ramped vanes

Experiments were performed at Mach 2 to assess the effectiveness of Ramped Vanes (RVs) in delaying the onset of shock-induced flow separation due to an axisymmetric compression corner. A cone–cylinder-flare model with a 24° flow deflection angle at the cylinder/flare juncture was selected for this study. A single row of RVs was placed 50 mm (10δ) upstream of the compression corner, to manipulate the incoming boundary layer and thereby bring alterations to the shock–boundary layer interaction flowfield. RVs with three sub-boundary layer heights viz. 1.4 mm, 2.1 mm and 2.8 mm (h/δRV=0.33, 0.50 and 0.67 respectively) were employed and their separation control performance was compared. Counter-rotating vortices originated from these devices, generating alternate upwash and downwash bands within the boundary layer. When the shorter devices were used, early and delayed onset of separation were noticed along the upwash and downwash regions respectively. However, as the devices grew taller, the increase in separation length along the upwash region seemed to diminish. Ultimately, when the tallest devices (h=2.8 mm) were introduced, the entire corrugated separation line was successfully pushed downstream of the uncontrolled separation line. Besides instigating these variations in the separation scale, the vortices also caused topological changes within the separation region. But, despite achieving a substantial delay in separation, the RVs were unable to shift the separation shock’s oscillations to higher frequencies.

Large-Eddy Simulation of Transonic Buffet Using Matrix-Free Discontinuous Galerkin Method

We present an implicit large-eddy simulation of transonic buffet over the OAT15A supercritical airfoil at Mach number 0.73, angle of attack 3.5 deg, and Reynolds number . The simulation is performed using a matrix-free discontinuous Galerkin (DG) method and a diagonally implicit Runge–Kutta scheme on graphics processor units. We propose a Jacobian-free Newton–Krylov method to solve nonlinear systems arising from the discretization of the Navier–Stokes equations. The method successfully predicts the buffet onset, the buffet frequency, and turbulence statistics owing to the high-order DG discretization and an efficient mesh refinement for the laminar and turbulent boundary layers. A number of physical phenomena present in the experiment are captured in our simulation, including periodical low-frequency oscillations of shock wave in the streamwise direction, strong shear layer detached from the shock wave due to shock-wave/boundary-layer interaction and small-scale structures broken down by the shear-layer instability in the transition region, and shock-induced flow separation. The pressure coefficient, the root mean square of the fluctuating pressure, and the streamwise range of the shock wave oscillation agree well with the experimental data. The results suggest that the proposed method can accurately predict the onset of turbulence and buffet phenomena at high Reynolds numbers without a subgrid scale model or a wall model.

A study on the flow physics of a supersonic partial admission impulse turbine for different tip clearances

Compared to traditional turbines, supersonic partial admission impulse (SPAI) turbines have significantly different flow physics owing to partial admission configuration and high supersonic flow at the rotor entrance. The objective of the present study is to reveal the intrinsic flow physics of an SPAI turbine for different rotor tip clearance dimensions. Numerical simulations were carried out based on the unsteady Reynolds-averaged Navier-Stokes (URANS) method in an SPAI turbine whose time-averaged relative Mach number at the rotor inlet was up to 2.1. The partial admission configuration led to a severer blockage in the hub region of the rotor entrance than that in the mid-span and tip regions. As a consequence, corner separation occurred near the leading edge of the rotor hub. The tip leakage flow was divided into two vortices. Due to the strong shock/tip leakage vortex interaction, the first vortex broke down and disappeared immediately. The second vortex originated from the leading edge shock foot, migrated downstream toward the mid-span (near the rotor trailing edge). It subsequently interacted with the trailing edge shock and exhibited an expansion process without experiencing a breakdown. Under the unsteady conditions, slight variations were observed in the shock train at the nozzle as the rotor passed. In contrast, due to the partial admission, the blade loading and the rotor leading edge shock experienced significant periodic variations. The rotor blade exhibited the maximum loading when the bulk flow originated from the nozzle flowed toward the rotor blade, whereas the minimum loading was achieved when the nozzle induced wake streamed toward the rotor blade. The leading edge shock/tip leakage vortex interaction occurred periodically, resulting in a periodic vortex breakdown. For the lowest efficiency condition, the highest total to static pressure ratio was achieved and the maximum intensity was exhibited for the leading edge shock of the rotor blade. This generated a significant loss due to the shock and the shock-induced flow separation.

Evaluation of reduced-order aeroelastic simulations for shock-dominated flows

Aeroelastic predictions of a thin panel in supersonic flow with a shock/boundary layer interaction (SBLI) are evaluated relative to wind tunnel experiments. Measured time histories of the panel displacement and velocity show stationary deflections as well as post-flutter oscillations for a range of shock impingement strengths. The fully coupled, computationally efficient aeroelastic modeling framework is formulated with a nonlinear structural reduced-order model and quasi-steady enriched piston theory aerodynamics. The aeroelastic model predicts the observed shape of the stationary responses with correlation coefficients larger than 0.8 for all SBLI conditions. In regards to dynamic responses, the model captures flutter onset as well as the post-flutter periodic oscillations for attached SBLI. The system’s sensitivity to different initial conditions and shock impingement location is also investigated for attached SBLI. Shock-induced flow separation emerges as the SBLI strength increases. For the strongest shock, the computations predict flutter but do not capture the frequency content of the measured post-flutter response. This discrepancy is tied to the breakdown of the quasi-steady flow assumption in piston theory with increasing flow separation and highlights the need for unsteady flow models.

On the unsteady dynamics of partially shrouded compressible jets

We experimentally investigate a partially shrouded sonic jet (a sonic free-jet shielded by a solid wall-extension on one side) exiting from a planar nozzle at two different nozzle pressure ratio ($\zeta=4$ and $5$). We experimentally show that the inherent jet unsteadiness from the shock-induced flow separation on the wall and the emitted noise in the far-field is strongly coupled through a series of experiments like high-speed schlieren, wall-static pressure, unsteady pressure spectra, and microphone measurements. The partially shrouded jet's lateral free expansion is also identified to be complicated, three-dimensional, and the produced noise is directional. The emitted acoustic pulses from the flapping-jet, the radiated noise from the shock-induced separation on the wall, and the shock-shear layer interaction on the other side of the wall are responsible for the generated acoustic disturbances. The non-uniform aeroacoustic forcing on the top and bottom portion of the partially wall-bounded jet shear layer leads to a self-sustained jet oscillation and a discrete sound emission. The vital features are identified through the proper orthogonal decomposition of high-speed schlieren images and supplemented by other measurements.

Topological modifications due to ramped vanes in a flare-induced shock–boundary layer interaction flowfield

Effect of ′ramped vane′-type vortex generators on a shock-induced flow separation in the vicinity of an axisymmetric compression corner was evaluated. Numerical simulations were performed at Mach 2 on a cone–cylinder–flare model with a flare angle of 24°. The undisturbed boundary layer thickness (δ) at the location of the compression corner was 5 mm. A single array of these vortex generators with a device height of 0.28δ (1.4 mm) was placed on the cylinder surface at different streamwise positions, viz. 5δ, 10δ and 15δ upstream of the compression corner, and their ability to manipulate the shock–boundary layer interaction flowfield was compared. The presence of these devices caused substantial changes in the interaction region and the separation bubble structure. The separation bubble transformed into a series of spade-shaped structures with pockets of attached flow in between them. The ramped vanes increased the separation length along the device centreline, but this effect was attenuated considerably, by bringing them closer to the interaction region. Moving the ramped vanes closer also had a collapsing effect on the spade-shaped structures, which simultaneously widened the attached flow zones in between them.

Investigation of shock dynamics in an axisymmetric inlet/isolator with attached boundary layers

Abstract

Control of Fin Shock Induced Flow Separation Using Vortex Generators

This work demonstrates the control of the shock-induced separated flow generated by a sharp fin placed on a cylindrical surface (fin-on-cylinder shock-wave/boundary-layer interaction) with an incident Mach 2.5 flow. Well-quantified distortions in the form of streamwise vortices are embedded into the boundary layer using single sub-boundary-layer vortex generators (VGs). At four device heights downstream from the VG trailing edge, the boundary layer exhibits a peak deficit in the streamwise velocity from the undistorted boundary layer of 60%. At 15 device heights, the boundary layer was energized to a peak surplus of 25% in the streamwise velocity. These two VG streamwise placements (four and 15 device heights) were employed to emphasize the impact of direct vortex interactions and boundary-layer energization on separation vortex control. It was found that having an energized boundary-layer incident on the fin-on-cylinder SBLI made very little change to the separation vortex size and the associated pressure distribution. However, placing the vortex generator much closer significantly mitigated the separation vortex size, and several azimuthal placement locations showed an elimination of the separation vortex. Cross-sectional velocity and vorticity fields showed a strong reduction in spanwise velocity and streamwise vorticity above the separation vortex in the successful configurations.

Experimental Investigation of Transonic Buffet Control on a Wing Airfoil Using Tangential Jet Blowing

The results of an experimental investigation of the suppression of shock-induced flow separation and the elimination of transonic buffet in transonic flow regimes using tangential jet blowing onto the upper surface of a supercritical wing airfoil are presented. The experimental investigation is performed in the T-112 wind tunnel of the Central Aerohydrodynamic Institute (TsAGI). The results obtained for the baseline configuration and those with jet blowing are analyzed and compared. They showed that the tangential jet blowing suppresses the shock-induced flow separation, increases the lift, and delays the onset of the buffet.

Shock-Induced Flow Separation in an Overexpanded Supersonic Planar Nozzle

Shock-induced flow separation in an overexpanded supersonic planar nozzle is investigated numerically by means of three-dimensional wall-modeled large-eddy simulations (LES). The objective of this study is to identify the origin of the low-frequency shock oscillations (LFO) and the associated side-loads generation in planar nozzles. The computational results are compared with the experimental data for validation. The promise of the near-wall LES modeling approach, adopted in this study, is supported by its satisfactory performance in correctly predicting the shock-induced flow separation and offering a major advantage of being 30–40 times faster than the wall-resolved LES counterpart, allowing thereby the capture of very–low-frequency shock oscillations with much better statistics convergence. The simulations bring clear evidence of the existence of broadband and energetically significant LFO in the vicinity of the separated shock, whose forward and backward movements are mainly driven by changes in the downstream flow conditions. The complex interactions between the backflow, the separation bubbles, and the large-scale turbulent structures developing in the shear-layer region strongly influence the shock unsteadiness, which in turn drives the LFO. A scenario of the LFO, confirming conclusions from earlier studies, is described in this work.