Showing papers on "Flow separation published in 1995"

Separated flow computations with the k-e-v2 model

Abstract: Tlirbulent separated flows over a backstep, in a plane diffuser and around a triangular cylinder, are computed with the k-e-v2 model. These provide examples of massive separation, of smooth separation, and of unsteady vortex shedding. It is shown that to accurately predict the time-averaged properties of bluff body flow, it is necessary to resolve the coherent vortex shedding. The near-wall treatment of the v2-/22 system of equations is able to cope with both the massive and smooth separations. Good agreement between experiment and prediction is found in all

Numerical/experimental study of a wingtip vortex in the near field

Abstract: The near-field behavior of a wingtip vortex flow has been studied computationally and experimentally in an interactive fashion. The computational approach involved using the method of artificial compressibility to solve the three-dimensional, incompressible, Navier-Stokes equations with experimentally determined boundary conditions and a modified Baldwin-Barth turbulence model. Inaccuracies caused by the finite difference technique, grid resolution, and turbulence modeling have been explored. The complete geometry case was computed using 1.5 million grid points and compared with mean velocity measurements on the suction side of the wing and in the near wake. Good agreement between the computed and measured flowfields has been obtained. The velocity distribution in the vortex core compares to within 3% of the experiment.

Details of the drag curve near the onset of vortex shedding

Abstract: A closer look at the drag curve for a circular cylinder near the onset of vortex shedding reveals that there is a sharp transition in the forces acting on a body moving through a fluid when it produces an unsteady wake. In this Letter results from high‐resolution computer simulations are presented to quantify the change in viscous drag, pressure drag, and base pressure coefficients.

High-resolution simulations of the flow around an impulsively started cylinder using vortex methods

Abstract: The development of a two-dimensional viscous incompressible flow generated from a circular cylinder impulsively started into rectilinear motion is studied computationally. An adaptative numerical scheme, based on vortex methods, is used to integrate the vorticity/velocity formulation of the Navier–Stokes equations for a wide range of Reynolds numbers (Re = 40 to 9500). A novel technique is implemented to resolve diffusion effects and enforce the no-slip boundary condition. The Biot–Savart law is employed to compute the velocities, thus eliminating the need for imposing the far-field boundary conditions. An efficient fast summation algorithm was implemented that allows a large number of computational elements, thus producing unprecedented high-resolution simulations. Results are compared to those from other theoretical, experimental and computational works and the relation between the unsteady vorticity field and the forces experienced by the body is discussed.

Mean flow and turbulence structure over fixed, two-dimensional dunes: implications for sediment transport and bedform stability

Abstract: Detailed measurements of flow velocity and its turbulent fluctuation were obtained over fixed, two-dimensional dunes in a laboratory channel. Laser Doppler anemometry was used to measure the downstream and vertical components of velocity at more than 1800 points over one dune wavelength. The density of the sampling grid allowed construction of a unique set of contour maps for all mean flow and turbulence parameters, which are assessed using higher moment measures and quadrant analysis. These flow field maps illustrate that: (1) the time-averaged downstream and vertical velocities agree well with previous studies of quasi-equilibrium flow over fixed and mobile bedforms and show a remarkable symmetry from crest to crest; (2) the maximum root-mean-square (RMS) of the downstream velocity values occur at and just downstream of flow reattachment and within the flow separation cell; (3) the maximum vertical RMS values occur within and above the zone of flow separation along the shear layer and this zone advects and diffuses downstream, extending almost to the next crest; (4) positive downstream skewness values occur within the separation cell, whereas positive vertical skewness values are restricted to the shear layer; (5) the highest Reynolds stresses are located within the zone of flow separation and along the shear layer; (6) high-magnitude, high-frequency quadrant-2 events (‘ejections’) are concentrated along the shear layer (Kelvin-Helmholtz instabilities) and dominate the contribution to the local Reynolds stress; and (7) high-magnitude, high-frequency quadrant-4 events occur bounding the separation zone, near reattachment and close to the dune crest, and are significant contributors to the local Reynolds stress at each location. These data demonstrate that the turbulence structure associated with dunes is controlled intrinsically by the formation, magnitude and downstream extent of the flow separation zone and resultant shear layer. Furthermore, the origin of dune-related macroturbulence lies in the dynamics of the shear layer rather than classical turbulent boundary layer bursting. The fluid dynamic distinction between dunes and ripples is reasoned to be linked to the velocity differential across the shear layer and hence the magnitude of the Kelvin-Helmholtz instabilities, which are both greater for dunes than ripples. These instabilities control the local flow and turbulence structure and dictate the modes of sediment entrainment and their transport rates.

Vortex shedding and shear-layer instability of wing at low-Reynolds numbers

Abstract: Flow patterns and characteristics of vortex shedding and shear-layer instability of a NACA 0012 cantilever wing are experimentally studied. Smoke-wire and surface oil-flow techniques are employed to visualize the flow patterns and evolution of vortex shedding. Hot-wire anemometers are used to characterize the frequency domain of the unsteady flow structures. Several characteristic flow modes are classified in the domain of chord Reynolds number and root angle of attack. Effects of the juncture and wing tip are discussed. Vortex shedding can be classified into four characteristic modes. Vortex shedding at low and high angles of attack are found to have different dominant mechanisms. Effects of the juncture and wing tip on the vortex shedding are discussed. Shear-layer instabilities are found to be closely related to the behaviors of the vortex shedding. Behaviors of the shear-layer instabilities can be traced back to the characteristics of the boundary layer on the suction surface of the airfoil.

Experimental investigation of flow through an asymmetric plane diffuser

Abstract: There is a need for experimental measurements in complex turbulent flows that originate from very well-defined initial conditions. Testing of large-eddy simulations and other higher-order computation schemes requires inlet boundary condition data that are not normally measured. The use of fully developed upstream conditions offers a solution to this dilemma so that the upstream conditions can be adequately computed at any level of sophistication. The plane diffuser experiment by Obi et al. (1993) has received a lot of attention because it has fully-developed inlet conditions and it includes separation from a smooth wall, subsequent reattachment and redevelopment of the downstream boundary layer. The objective of this study is to provide careful qualification and detailed measurements in a recreation of the Obi experiment. The work will include extensive documentation of the flow two-dimensionality and detailed measurements required for testing of flow computations.

The onset of separation in neutral, turbulent flow over hills

Abstract: The onset of separation in turbulent, neutrally stratified, boundary-layer flow over hills is considered. Since the flows are fully turbulent, the occurrence of intermittent separation, in the sense of any reversal of near surface flow, will depend strongly on the detailed structure and behaviour of the turbulent eddies. Very little is known about such intermittent separation and the phenomenon cannot be studied with numerical models employing standard turbulence closures; eddy-resolving models are required. Therefore, here, as elsewhere in the literature, the arguably less physically significant process of mean flow separation is studied. Numerical simulations of flow over idealised two- and three-dimensional hills are examined in detail to determine the lowest slope,Θ crit, for which the mean flow separates. Previous work has identified this critical slope as that required to produce a zero surface stress somewhere over the hill. This criterion, when a mixing-length turbulence closure is applied, reduces to requiring the near-surface vertical velocity shear to vanish at some point on the hill's surface. By applying results from a recent linear analysis for the flow perturbations to this condition, a new expression forΘ crit is obtained. The expression is approximate but its relative simplicity makes it practically applicable without the need for use of a computer or for detailed mapping of the hill. The approach suggested differs from previous ones in that it applies linear results to a non-linear expression for the surface stress. In the past, a linear expression for the surface stress has been used. The proposed expression forΘ crit leads to critical angles that are about twice previous predictions. It is shown that the present expression gives good agreement with the numerical results presented here, as well as with other numerical and experimental results. It is also consistent with atmospheric observations.

Reynolds number effect on the skin friction in separated flows behind a backward-facing step

Abstract: Results from the present study show that the skin-friction coefficient decreases as the Reynolds number,Re h , increases in the following manner, C f ,min=−019Re h −1/2 The -1/2 power relationship deduced from the correlationC f,min vsRe h indicates laminar like behavior which is consistent with the findings of Adams et al (1984) Clauser's method, which is frequently used for the determination of the wall shear stress, leads to erroneous results when applied to the velocity measurements obtained in the near field of reattaching flows (and many other wall-bounded nonequilibrium flows) Direct measurements of theC f using the LOI technique give higher values than those obtained by the classical techniques The normalized mean velocity on the wall coordinates violates the universal law-of-the-wall in the near field of reattaching flows

Effect of concentrated wall suction on a turbulent boundary layer

Abstract: The effect of suction, applied through a short porous wall strip, on a low Reynolds number self‐preserving turbulent boundary layer has been quantified by measuring the local wall shear stress and the main Reynolds stresses downstream of the strip. When the suction rate is sufficiently high, pseudo‐relaminarization occurs almost immediately downstream of the strip. Farther downstream, transition occurs followed by a slow return to a fully turbulent self‐preserving state. During relaminarization, the measured skin friction coefficient cf falls below the level corresponding to the no suction value, reaching a minimum where transition begins. An empirical cf distribution is proposed that groups together results obtained at different streamwise stations and different suction rates. Of all the measured Reynolds stresses, the longitudinal turbulence intensity recovers relatively quickly from the change in boundary conditions while the wall‐normal turbulence intensity and the Reynolds shear stress are significan...

Compressible cavity flow oscillation due to shear layer instabilities and pressure feedback

Abstract: A computational analysis was performed on a compressible flow oscillation due to shear layer instabilities over a cavity and pressure feedback in a cavity of length-to-depth ratio 3 at Mach 1.5 and 2.5. The mass-averaged NavierStokes equations were solved. Turbulence closure was achieved using a k-u> model with compressibility corrections. Self-sustained oscillations were produced. Negative form drag coefficient was observed within an oscillatory cycle due to mass ejection from the cavity near the trailing edge and vortex production near the leading edge. The shock wave-expansion wave interaction patterns, modes of the oscillation, sound pressure level, and time-averaged surface pressure were compared with experimental results of previous investigations and good agreement was achieved, particularly the time-averaged pressure. The prediction showed a marked improvement over earlier analysis.

Turbulent and Laminar Drag of Superfluid Helium on an Oscillating Microsphere.

Abstract: A magnetic sphere (radius 100 μm), levitating inside a superconducting niobium capacitor, is immersed into superfluid helium. Vertical oscillations of the sphere can be excited and detected. At resonance we measure the velocity amplitude as a function of the driving force between 0.35 and 2.2 K. In the linear regime (laminar flow) the drag is given by Stokes' solution above 1.1 K and by ballistic roton and phonon drag below 0.7 K. At larger velocities we find a sharp transition to turbulent drag which varies with the square of the velocity above a temperature independent threshold velocity.

The structure of tip clearance flow in axial flow compressors

Abstract: Detailed measurements of the flow field in the tip region of an axial flow compressor rotor were carried out using a rotating five-hole probe. The axial, tangential, and radial components of relative velocity, as well as the static and stagnation pressures, were obtained at two axial locations, one at the rotor trailing edge, the other downstream of the rotor. The measurements were taken up to about 26 percent of the blade span from the blade tip. The data are interpreted to understand the complex nature of the flow in the tip region, which involves the interaction of the tip leakage flow, the annulus wall boundary layer and the blade wake. The experimental data show that the leakage jet does not roll up into a vortex. The leakage jet exiting from the tip gap is of high velocity and mixes quickly with the mainstream, producing intense shearing and flow separation. There are substantial differences in the structure of tip clearance observed in cascades and rotors.

Turbulent vortex breakdown

Abstract: Reported herein is a cone‐shaped turbulent vortex breakdown in noncavitating swirling flows at high Reynolds numbers in a slightly diverging cylindrical tube. The turbulent conical form is in addition to the well‐known double‐helix, spiral, and nearly axisymmetric or ‘‘bubble’’‐type breakdowns.

Influence of a strong adverse pressure gradient on the turbulent structure in a boundary layer

Abstract: A comparison between the turbulent structures found in a zero pressure gradient boundary layer and a boundary layer subjected to a strong adverse pressure gradient is presented. The pressure gradient reverses the direction of the dominant turbulent diffusion, resulting in considerable turbulent transport towards the wall. Two‐point space–time correlations and the invariants show that this reduces the anisotropy in the near wall region and indicate an important reflection of the turbulent motion from the wall back into the outer layer. This is verified by a quadrant analysis [Lu and Willmarth, J. Fluid Mech. 60, 481 (1973)] which demonstrates that the strong events near the wall are totally dominated by motions in the first and fourth quadrants.

An experimental study of a three-dimensional pressure-driven turbulent boundary layer

Abstract: A three-dimensional, pressure-driven turbulent boundary layer created by an idealized wing-body junction flow was studied experimentally. The data presented include time-mean static pressure and directly measured skin-friction magnitude on the wall. The mean velocity and all Reynolds stresses from a three-velocity-component fibre-optic laser-Doppler anemometer are presented at several stations along a line determined by the mean velocity vector component parallel to the wall in the layer where the u 2 kinematic normal stress is maximum (normal-stress coordinate system). This line was selected by intuitively reasoning that overlap of the near-wall flow and outer-region flow occurs at the location where u 2 is maximum. Along this line the flow is subjected to a strong crossflow pressure gradient, which changes sign for the downstream stations. The shear-stress vector direction in the flow lags behind the flow gradient vector direction. The flow studied here differs from many other experimentally examined three-dimensional flows in that the mean flow variables depend on three spatial axes rather than two axes, such as flows in which the three-dimensionality of the flow has been generated either by a rotating cylinder or by a pressure gradient in one direction only throughout the flow.

ONERA Three-Dimensional Icing Model

Abstract: A three-dimensional icing model has been developed at ONERA to calculate ice accretion shapes for aerodynamic components that can not be predicted using conventional two-dimensional codes. It is described, emphasizing the original parts with respect to the two-dimensional existing models. The model includes Euler inviscid flow calculation. Droplet trajectories are calculated in a three-dimensional grid. The remesh on the leading edge is adapted to follow aerodynamics singularities. The boundary layer is calculated using a mixing length formulation to model the wall roughness influence on convective heat transfer. Runback paths are integrated. The heat balance is calculated in a grid created along the runback paths. The domain of validity of the three-dimensional icing code is described; compared with the two-dimensional model this domain is wider, especially for high speeds. The three-dimensional model is shown to simulate well a uniform ice deposit on a three-dimensional rotor blade tip. Then, a comparison of the three- and two-dimensional codes on an infinite swept wing shows that the corrected two-dimensional code predicts the catch efficiency but not the ice shape. Finally, it is shown that the continuum flux hypothesis prevents the three-dimensional model from simulating correctly the "lobster tail" ice shape (nonuniform ice deposit).

Numerical Investigation of Supersonic Flows Around a Spiked Blunt Body

Abstract: In supersonic flow, a spike attached to the nose reduces the drag of a blunt body. In this paper, supersonic flows around a spiked blunt body are numerically simulated to examine the effects of the spike length, Mach number, and angle of attack. Three-dimensional thin-layer compressible Navier-Stokes equations are solved using a highresolution upwind scheme with LU-ADI time-integration algorithm. The computed results show that the drag of the spiked blunt body is significantly influenced by the spike length, Mach number, and angle of attack. Scales of the separated region are not significantly influenced by the freestream Mach number. For the spiked blunt body at angle of attack, the flowfield becomes complex with spiral flows. The computed results are in reasonable agreement with experimental data.

Comparative measurements in the canonical boundary layer at Reδ2≤6×104 on the wall of the German–Dutch windtunnel

Abstract: Mean velocity and Reynolds‐stress profiles were measured in the incompressible turbulent boundary layer with zero pressure gradient on the aerodynamically smooth sidewall of the German–Dutch windtunnel. Data were taken at Reynolds numbers Reδ2, based on momentum thickness δ2 of 2×104, 4×104, and 6×104 by means of four different types of hot‐wire probes (three‐wire probes, X wire, and normal‐wire probes). There are also measurements of skin friction and of spectra. The data compare well with the few available other measurements, and all profiles show independence of Reynolds number in the outer region of the boundary layer when plotted against y/Δ, where Δ is the Rotta–Clauser length.

Unsteady flow about a sphere at low to moderate Reynolds number. Part 2. Accelerated motion

Abstract: A full numerical simulation based on spectral methods is used to investigate linearly accelerating and decelerating flows past a rigid sphere. Although flow separation does not occur at Reynolds numbers below 20 for a steady flow, in the linearly decelerating flow separation is observed at much lower Reynolds numbers with complete detachment of vorticity possible in certain cases. The existence of a large recirculation region contributes to the result that a negative viscous force on the sphere is possible. The contribution of the pressure to the force includes a component that is well described by the inviscid added-mass term in both the accelerating and decelerating cases. The force on the sphere is found in general to initially decay in a power law manner after acceleration or deceleration ends followed by rapid convergence at later times to the steady state. For the cases examined this convergence is found to be exponential except for those in which the sphere is brought to rest in which case the convergence remains algebraic. This includes the special case of an infinite acceleration or deceleration where the free stream velocity is impulsively changed.

Boundary Layer Development in Axial Compressors and Turbines: Part 1 of 4 — Composite Picture

Abstract: Comprehensive experiments and computational analyses were conducted to understand boundary layer development on airfoil surfaces in multistage, axial-flow compressors and LP turbines. The tests were run over a broad range of Reynolds numbers and loading levels in large, low-speed research facilities which simulate the relevant aerodynamic features of modern engine components. Measurements of boundary layer characteristics were obtained by using arrays of densely packed, hot-film gauges mounted on airfoil surfaces and by making boundary layer surveys with hot wire probes. Computational predictions were made using both steady flow codes and an unsteady flow code. This is the first time that time-resolved boundary layer measurements and detailed comparisons of measured data with predictions of boundary layer codes have been reported for multistage compressor and turbine blading.Part 1 of this paper draws a composite picture of boundary layer development in turbomachinery based upon a synthesis of all of our experimental findings for the compressor and turbine. Parts 2 and 3 present the experimental results for the compressor and turbine, respectively. Part 4 presents computational analyses and discusses comparisons with experimental data.For both compressor and turbine blading, the experimental results show large extents of laminar and transitional flow on the suction surface of embedded stages, with the boundary layer generally developing along two distinct but coupled paths. One path lies approximately under the wake trajectory while the other lies between wakes. Along both paths the boundary layer clearly goes from laminar to transitional to turbulent. The wake path and the non-wake path are coupled by a calmed region which, being generated by turbulent spots produced in the wake path, is effective in suppressing flow separation and delaying transition in the non-wake path. The location and strength of the various regions within the paths, such as wake-induced transitional and turbulent strips, vary with Reynolds number, loading level and turbulence intensity. On the pressure surface, transition takes place near the leading edge for the blading tested. For both surfaces, bypass transition and separated-flow transition were observed. Classical Tollmien-Schlichting transition did not play a significant role. Comparisons of embedded and first-stage results were also made to assess the relevance of applying single-stage and cascade studies to the multistage environment.Although doing well under certain conditions, the codes in general could not adequately predict the onset and extent of transition in regions affected by calming. However, assessments are made to guide designers in using current predictive schemes to compute boundary layer features and obtain reasonable loss predictions.Copyright © 1995 by ASME

Simulations of the unsteady separated flow past a normal flat plate

Abstract: Well-resolved two-dimensional numerical simulations of the unsteady separated flow past a normal flat plate at low Reynolds numbers have been performed using a fractional step procedure with high-order spatial discretization. A fifth-order upwind-biased scheme is used for the convective terms and the diffusive terms are represented by a fourth-order central difference scheme. The pressure Poisson equation is solved using a direct method based on eigenvalue decomposition of the coefficient matrix. A systematic study of the flow has been conducted with high temporal and spatial resolutions for a series of Reynolds numbers. The interactions of the vortices shed form the shear layers in the near-and far-wake regions are studied. For Reynolds numbers less than 250 the vortices are observed to convect parallel to the freestream. However, at higher Reynolds numbers (500 and 1000), complex interactions including vortex pairing, tearing and deformations are seen to occur in the far-wake region. Values of the drag coefficient and the wake closure length are presented and compared with previous experimental and numerical studies.

Real-time prediction of unsteady aerodynamics: Application for aircraft control and manoeuvrability enhancement

Abstract: The capability to control unsteady separated flow fields could dramatically enhance aircraft agility. To enable control, however, real-time prediction of these flow fields over a broad parameter range must be realized. The present work describes real-time predictions of three-dimensional unsteady separated flow fields and aerodynamic coefficients using neural networks. Unsteady surface-pressure readings were obtained from an airfoil pitched at a constant rate through the static stall angle. All data sets were comprised of 15 simultaneously acquired pressure records and one pitch angle record. Five such records and the associated pitch angle histories were used to train the neural network using a time-series algorithm. Post-training, the input to the network was the pitch angle (/spl alpha/), the angular velocity (d/spl alpha//dt), and the initial 15 recorded surface pressures at time (t/sub 0/). Subsequently, the time (t+/spl Delta/t) network predictions, for each of the surface pressures, were fed back as the input to the network throughout the pitch history. The results indicated that the neural network accurately predicted the unsteady separated flow fields as well as the aerodynamic coefficients to within 5% of the experimental data. Consistent results were obtained both for the training set as well as for generalization to both other constant pitch rates and to sinusoidal pitch motions. The results clearly indicated that the neural-network model could predict the unsteady surface-pressure distributions and aerodynamic coefficients based solely on angle of attack information. The capability for real-time prediction of both unsteady separated flow fields and aerodynamic coefficients across a wide range of parameters in turn provides a critical step towards the development of control systems targeted at exploiting unsteady aerodynamics for aircraft manoeuvrability enhancement. >