Determination of complete velocity gradient tensor by using cinematographic stereoscopic PIV in a turbulent jet

Cinematographic stereoscopic PIV measurements were performed in the far field of an axisymmetric co-flowing turbulent round jet (Re _T ≈ 150, where Re _T is the Reynolds number based on Taylor micro scale) to resolve small and intermediate scales of turbulence. The time-resolved three-component PIV measurements were performed in a plane normal to the axis of the jet and the data were converted to quasi-instantaneous three-dimensional (volumetric) data by using Taylor’s hypothesis. The availability of the quasi-three-dimensional data enabled the computation of all nine components of the velocity gradient tensor over a volume. The use of Taylor’s hypothesis was validated by performing a separate set of time-resolved two component “side-view” PIV measurements in a plane along the jet axis. Probability density distributions of the velocity gradients computed using Taylor’s hypothesis show good agreement with those computed directly with the spatially resolved data. The overall spatial structure of the gradients computed directly exhibits excellent similarity with that computed using Taylor’s hypothesis. The accuracy of the velocity gradients computed from the pseudo-volume was assessed by computing the divergence error in the flow field. The root mean square (rms) of the divergence error relative to the magnitude of the velocity gradient tensor was found to be 0.25, which is consistent with results based on other gradient measurement techniques. The velocity gradients, vorticity components and mean dissipation in the self-similar far field of the jet were found to satisfy the axisymmetric isotropy conditions. The divergence error present in the data is attributed to the intrinsic uncertainty associated with performing stereoscopic PIV measurements and not to the use of Taylor’s hypothesis. The divergence error in the data is found to affect areas of low gradient values and manifests as nonphysical values for quantities like the normalized eigenvalues of the strain-rate tensor. However, the high gradients are less affected by the divergence error and so it can be inferred that structural features of regions of intense vorticity and dissipation will be faithfully rendered.     

The effects of resolution and noise on kinematic features of fine-scale turbulence

The effect of spatial resolution and experimental noise on the kinematic fine-scale features in shear flow turbulence is investigated by means of comparing numerical and experimental data. A direct numerical simulation (DNS) of a nominally two-dimensional planar mixing layer is mean filtered onto a uniform Cartesian grid at four different, progressively coarser, spatial resolutions. Spatial gradients are then calculated using a simple second-order scheme that is commonly used in experimental studies in order to make direct comparisons between the numerical and previously obtained experimental data. As expected, consistent with other studies, it is found that reduction of spatial resolution greatly reduces the frequency of high magnitude velocity gradients and thereby reduces the intermittency of the scalar analogues to strain (dissipation) and rotation (enstrophy). There is also an increase in the distances over which dissipation and enstrophy are spatially coherent in physical space as the resolution is coarsened, although these distances remain a constant number of grid points, suggesting that the data follow the applied filter. This reduction of intermittency is also observed in the eigenvalues of the strain-rate tensor as spatial resolution is reduced. The quantity with which these eigenvalues is normalised is shown to be extremely important as fine-scale quantities, such as the Kolmogorov length scale, are showed to change with different spatial resolution. This leads to a slight change in the modal values for these eigenvalues when normalised by the local Kolmogorov scale, which is not observed when they are normalised by large-scale, resolution-independent quantities. The interaction between strain and rotation is examined by means of the joint probability density function (pdf) between the second and third invariants of the characteristic equation of the velocity gradient tensor, Q and R respectively and by the alignments between the eigenvectors of the strain-rate tensor and the vorticity vector. Gaussian noise is shown to increase the divergence error of a dataset and subsequently affect both the Q–R joint pdf and the magnitude of the alignment cosines. The experimental datasets are showed to behave qualitatively similarly to the numerical datasets to which Gaussian noise has been added, confirming the importance of understanding the limitations of coarsely resolved, noisy experimental data.     

Stereoscopic PIV measurements of a turbulent boundary layer with a large spatial dynamic range

The flow in a streamwise/wall-normal plane of a turbulent boundary layer at moderate Reynolds number (Re _θ = 2,200) is characterized using two stereo PIV systems just overlapping in the streamwise direction. The aim is to generate SPIV data for near-wall turbulence with enough spatial dynamic range to resolve most of the coherent structures present in the flow and to facilitate future comparisons with direct numerical simulations. This is made possibly through the use of four cameras with large CCD arrays (4,008 px × 2,672 px) and through a rigorous experimental procedure designed to minimize the impact of measurement noise on the resolution of the small scales. For the first time, both a large field of view [S _x ; S _y ] = [2.6δ; 0.75δ] and a high spatial resolution (with an interrogation window size of 13.6⁺) have been achieved. The quality of the data is assessed through an analysis of some of the statistical results such as the mean velocity profile, the rms and the PDF of the fluctuations, and the power spectra.     

The accuracy of tomographic particle image velocimetry for measurements of a turbulent boundary layer

To investigate the accuracy of tomographic particle image velocimetry (Tomo-PIV) for turbulent boundary layer measurements, a series of synthetic image-based simulations and practical experiments are performed on a high Reynolds number turbulent boundary layer at Re_θ = 7,800. Two different approaches to Tomo-PIV are examined using a full-volume slab measurement and a thin-volume “fat” light sheet approach. Tomographic reconstruction is performed using both the standard MART technique and the more efficient MLOS-SMART approach, showing a 10-time increase in processing speed. Random and bias errors are quantified under the influence of the near-wall velocity gradient, reconstruction method, ghost particles, seeding density and volume thickness, using synthetic images. Experimental Tomo-PIV results are compared with hot-wire measurements and errors are examined in terms of the measured mean and fluctuating profiles, probability density functions of the fluctuations, distributions of fluctuating divergence through the volume and velocity power spectra. Velocity gradients have a large effect on errors near the wall and also increase the errors associated with ghost particles, which convect at mean velocities through the volume thickness. Tomo-PIV provides accurate experimental measurements at low wave numbers; however, reconstruction introduces high noise levels that reduces the effective spatial resolution. A thinner volume is shown to provide a higher measurement accuracy at the expense of the measurement domain, albeit still at a lower effective spatial resolution than planar and Stereo-PIV.     

Dissipation rate estimation from PIV in zero-mean isotropic turbulence

Measuring the turbulent kinetic energy dissipation rate in an enclosed turbulence chamber that produces zero-mean flow is an experimental challenge. Traditional single-point dissipation rate measurement techniques are not applicable to flows with zero-mean velocity. Particle image velocimetry (PIV) affords calculation of the spatial derivative as well as the use of multi-point statistics to determine the dissipation rate. However, there is no consensus in the literature as to the best method to obtain dissipation rates from PIV measurements in such flows. We apply PIV in an enclosed zero-mean turbulent flow chamber and investigate five methods for dissipation rate estimation. We examine the influence of the PIV interrogation cell size on the performance of different dissipation rate estimation methods and evaluate correction factors that account for errors related to measurement uncertainty, finite spatial resolution, and low Reynolds number effects. We find the Re _λ corrected, second-order, longitudinal velocity structure function method to be the most robust method to estimate the dissipation rate in our zero-mean, gaseous flow system.     

Unsteady PTV velocity field past an airfoil solved with DNS: Part 1. Algorithm of hybrid simulation and hybrid velocity field at <Emphasis Type="Italic">Re</Emphasis> ≈ 10<Superscript>3</Superscript>

We develop a hybrid unsteady-flow simulation technique combining direct numerical simulation (DNS) and particle tracking velocimetry (PTV) and demonstrate its capabilities by investigating flows past an airfoil. We rectify instantaneous PTV velocity fields in a least-squares sense so that they satisfy the equation of continuity, and feed them to the DNS by equating the computational time step with the frame rate of the time-resolved PTV system. As a result, we can reconstruct unsteady velocity fields that satisfy the governing equations based on experimental data, with the resolution comparable to numerical simulation. In addition, unsteady pressure distribution can be solved simultaneously. In this study, particle velocities are acquired on a laser-light sheet in a water tunnel, and unsteady flow fields are reconstructed with the hybrid algorithm solving the incompressible Navier–Stokes equations in two dimensions. By performing the hybrid simulation, we investigate nominally two-dimensional flows past the NACA0012 airfoil at low Reynolds numbers. In part 1, we introduce the algorithm of the proposed technique and discuss the characteristics of hybrid velocity fields. In particular, we focus on a vortex shedding phenomenon under a deep stall condition (α = 15°) at Reynolds numbers of Re = 1000 and 1300, and compare the hybrid velocity fields with those computed with two-dimensional DNS. In part 2, the extension to higher Reynolds numbers is considered. The accuracy of the hybrid simulation is evaluated by comparing with independent experimental results at various angles of attack and Reynolds numbers up to Re = 10⁴. The capabilities of the hybrid simulation are also compared with two-dimensional unsteady Reynolds-Averaged Navier–Stokes (URANS) solutions in part 2. In the first part of these twin papers, we demonstrate that the hybrid velocity field approaches the PTV velocity field over time. We find that intensive alternate vortex shedding past the airfoil, which is predicted by the two-dimensional DNS, is substantially suppressed in the hybrid simulation and the resultant flow field is similar to the PTV velocity field, which is projection of the three-dimensional velocity field on the streamwise plane. We attempt to identify the motion that originates three-dimensional flow patterns by highlighting the deviation of the PTV velocity field from the two-dimensional governing equations at each snapshot. The results indicate that the intensive spots of the deviation appear in the regions in which three-dimensional instabilities are induced in the shear layer separated from the pressure side.     

On the accuracy of simultaneously measuring velocity component statistics in turbulent wall flows with arrays of three or four hot-wire sensors

A highly resolved turbulent channel flow direct numerical simulation (DNS) with Re _τ = 200 has been used to investigate the ability of probes made up of arrays of three or four hot-wire sensors to simultaneously and accurately measure statistics of all three velocity components in turbulent wall flows. Various virtual sensor arrangements have been tested in order to study the effects of position, number of sensors and spatial resolution on the measurements. First, the effective cooling velocity was determined for each sensor of an idealized probe, where the influence of the velocity component tangential to the sensors and flow blockage by the presence of the probe are neglected. Then, simulating the response of the virtual probes to obtain the effective velocities cooling the sensors, velocity component statistics have been calculated neglecting the velocity gradients over the probe sensing area. A strong influence of both mean and fluctuation velocity gradients on measurement accuracy was found. A new three-sensor array configuration designed to minimize the influence of the velocity gradients is proposed, and its accuracy is compared to two-sensor X- and V-array configurations.     

Scalar turbulence model investigation with variable turbulent Prandtl number in heated jets and diffusion flames

Traditional turbulence models using constant turbulent Prandtl number fail to predict the experimentally observed anisotropies in the thermal eddy diffusivity and thermal turbulent intensity fields. Accurate predictions depend strongly on the turbulence model employed. Consequently, the objective of this paper is to assess the performance of turbulence model with variable turbulent Prandtl number in predicting of thermal and scalar fields quantities. The model is applied to axisymmetric turbulent round jet with variable density and in turbulent hydrogen diffusion flames using the flamelet concept. The k − ɛ turbulence model is used in conjunction with thermal field; the model involves solving supplemental scalar equations for the temperature variance and its dissipation rate. The model predictions are compared with available experimental data for the purpose of validating model. In reacting cases, velocity and scalar (including temperature and mass fractions) predictions agree relatively well in the near field of the investigated diluted hydrogen flames.     

High-resolution imaging of turbulence structures in jet flames and non-reacting jets with laser Rayleigh scattering

A comparative study of the length scales and morphology of dissipation fields in turbulent jet flames and non-reacting jets provides a quantitative analysis of the effects of heat release on the fine-scale structure of turbulent mixing. Planar laser Rayleigh scattering is used for highly resolved measurements of the thermal and scalar dissipation in the near fields of CH₄/H₂/N₂ jet flames (Re _d  = 15,200 and 22,800) and non-reacting propane jets (Re _d  = 7,200–21,700), respectively. Heat release increases the dissipation cutoff length scales in the reaction zone of the flames such that they are significantly larger than the cutoff scales of non-reacting jets with comparable jet exit Reynolds numbers. Fine-scale anisotropy is enhanced in the reaction zone. At x/d = 10, the peaks of the dissipation angle PDFs in the Re _d  = 15,200 and 22,800 jet flames exceed those of non-reacting jets with corresponding jet exit Reynolds numbers by factors of 2.3 and 1.8, respectively. Heat release significantly reduces the dissipation layer curvature in the reaction zone and in the low-temperature periphery of the jet flames. These results suggest that the reaction zone shields the outer regions of the jet flame from the highly turbulent flow closer to the jet axis.     

Tailoring turbulence with an active grid

Using an active grid in a wind tunnel, we generate homogeneous shear turbulence and initiate turbulent boundary layers with adjustable properties. Homogeneous shear turbulence is characterized by a constant gradient of the mean velocity and a constant turbulence intensity. It is the simplest anisotropic turbulent flow thinkable, and it is generated traditionally by equipping a wind tunnel with screens which have a varying transparency and flow straighteners. This is not done easily, and the reachable turbulence levels are modest. We describe a new technique for generating homogeneous shear turbulence using an active grid only. Our active grid consists of a grid of rods with attached vanes which can be rotated by servo motors. We control the grid by prescribing the time-dependent angle of each axis. We tune the vertical transparency profile of the grid by setting appropriate angles of each rod such as to generate a uniform velocity gradient, and set the rods in flapping motion around these angles to tailor the turbulence intensity. The Taylor Reynolds number reached was R _λ = 870, the shear rate S = ∂U/∂y = 9.2 s⁻¹, the nondimensional shear parameter S ^*≡ Sq ²/ε = 12 and u = 1.4 ms⁻¹. As a further application of this idea we demonstrate the generation of a simulated atmospheric boundary layer in a wind tunnel which has tunable properties. This method offers a great advantage over the traditional one, in which vortex-generating structures need to be placed in the wind tunnel to initiate a fat boundary layer.     

A correction method for measuring turbulence kinetic energy dissipation rate by PIV

The accuracy of turbulent kinetic energy (TKE) dissipation rate measured by PIV is studied. The critical issue for PIV-based dissipation measurements is the strong dependency on the spatial resolution, Δx, as reported by Saarenrinne and Piirto (Exp Fluids Suppl:S300–S307, 2000). When the PIV spacing is larger than the Kolmogorov scale, η, the dissipation is underestimated because the small scale fluctuations are filtered. For the case of Δx smaller than the Kolmogorov scale, the error rapidly increases due to noise. We introduce a correction method to eliminate the dominant error for the small Δx case. The correction method is validated by using a novel PIV benchmark, random Oseen vortices synthetic image test (ROST), in which quasi-turbulence is generated by randomly superposing multiple Oseen vortices. The error of the measured dissipation can be more than 1,000% of the analytical dissipation for the small Δx case, while the dissipation rate is underestimated for the large Δx case. Though the correction method does not correct the underestimate due to the low resolution, the dissipation was accurately obtained within a few percent of the true value by using the correction method for the optimal resolution of η/10 < Δx < η/2.     

Oscillating flow in a 2-D diffuser

Separating oscillating flows in an internal, adverse pressure gradient geometry are studied experimentally. Simultaneous velocity and pressure measurements demonstrate that the minor losses associated with oscillating flow in an adverse pressure gradient geometry can be smaller or larger than those for steady flow. Separation is found to begin high in the diffuser and propagate downward. The flow is able to remain attached further into the diffuser with larger Reynolds numbers, small displacement amplitudes, and smaller diffuser angles. The extent of separation grows with L ₀/h. The minor losses grow with increasing displacement amplitude in the measured range 10 < L ₀/h < 40. Losses decrease with increasing Re _δ in the measured range of 380 < Re _δ < 740. It is found that the losses increase with increasing diffuser angle over the measured range of 12° < θ < 30°. The nondimensional acoustic power dissipation increases with Reynolds number in the measured range and decreases with displacement amplitude.     

Flow field analysis of a turbulent boundary layer over a riblet surface

The near-wall flow structures of a turbulent boundary layer over a riblet surface with semi-circular grooves were investigated experimentally for the cases of drag decreasing (s ⁺=25.2) and drag increasing (s ⁺=40.6). One thousand instantaneous velocity fields over riblets were measured using the velocity field measurement technique and compared with those above a smooth flat plate. The field of view was 6.75 × 6.75 mm² in physical dimension, containing two grooves. Those instantaneous velocity fields were ensemble averaged to get turbulent statistics including turbulent intensities and turbulent kinetic energy. To see the global flow structure qualitatively, flow visualization was also carried out using the synchronized smoke-wire technique under the same experimental conditions. For the case of drag decreasing (s ⁺=25.2), most of the streamwise vortices stay above the riblets, interacting with the riblet tips frequently. The riblet tips impede the spanwise movement of the streamwise vortices and induce secondary vortices. The normalized rms velocity fluctuations and turbulent kinetic energy are small near the riblet surface, compared with those over a smooth flat plate. Inside the riblet valleys, these are sufficiently small that the increased wetted surface area of the riblets can be compensated. In addition, in the outer region (y ⁺ > 30), these values are almost equal to or slightly smaller than those for the smooth plate. For the case of drag increasing (s ⁺=40.6), however, most of the streamwise vortices stay inside the riblet valleys and contact directly with the riblet surface. The high-speed down-wash flow penetrating into the riblet valley interacts actively with the wetted riblet surface and increases the skin friction. The rms velocity fluctuations and turbulent kinetic energy have larger values compared with those over a smooth flat plate. Received: 24 March 1999/Accepted: 10 March 2000     

Self-sustained oscillations of turbulent flows over an open cavity

The mechanism of self-sustained oscillations in laminar cavity flows has been well characterized; however, the occurrence of self-sustained oscillations in turbulent cavity flows has only previously been characterized by direct observation of flows. Here, the quantitative characteristics of vortical structures in turbulent flows over an open cavity were determined, and then statistical properties were examined for evidence of self-sustained oscillations. Specifically, instantaneous velocity fields were measured using PIV and wall pressure fluctuations were determined from microphone data. Cavity geometries of L/D = 1 and 2, where L and D are the length and depth of the cavity, respectively, were used under conditions where the incoming boundary layer was turbulent at Re _θ  = 830. Statistical analyses were applied based on the instantaneous velocity fields of PIV data. The spatial distributions of vertical velocity correlations (v–v) showed alternating patterns that reflect the organized nature of the large-scale vortical structures corresponding to the modes of N = 2 for L/D = 1 and N = 3 for L/D = 2. These values were consistent with the numbers of vortical structures obtained from a modified version of Rossiter’s equation. Furthermore the numbers of vortical structures determined in the statistical analyses were consistently observed in instantaneous distributions of the swirling strength (λ _ci). The incoming turbulent boundary layer can give rise to the formation of large-scale vortical structures responsible for self-sustained oscillations.     

Approach to local axisymmetry in a turbulent cylinder wake

The objective of this experimental study is to characterise the small-scale turbulence in the intermediate wake of a circular cylinder using measured mean-squared velocity gradients. Seven of the twelve terms which feature in ε, the mean dissipation rate of the turbulent kinetic energy, were measured throughout the intermediate wake at a Reynolds number of Re _d  ≈ 3000 based on the cylinder diameter (d). Earlier measurements of the nine major terms of ε by Browne et al. (J Fluid Mech 179: 307–326 1987) at a downstream distance (x) of x = 420d and Re _d  ≈ 1170 are also used. Whilst departures from local isotropy are significant at all locations in the wake, local axisymmetry of the small-scale turbulence with respect to the mean flow direction is first satisfied approximately at x = 40d. The approach towards local axisymmetry is discussed in some detail in the context of the relative values of the mean-squared velocity gradients. The data also indicate that axisymmetry is approximately satisfied by the large scales at x/d ≥ 40, suggesting that the characteristics of the small scales reflect to a major extent those of the large scales. Nevertheless, the far-wake data of Browne et al. (1987) show a discernible departure from axisymmetry for both small and large scales.     

On the spatial resolution of velocity and velocity gradient-based turbulence statistics measured with multi-sensor hot-wire probes

A highly resolved turbulent channel flow direct numerical simulation with Re _τ = 200 has been used to investigate the ability of 12-sensor hot-wire probes to accurately measure velocity and velocity gradient based turbulence statistics. Various virtual sensor separations have been tested in order to study the effects of spatial resolution on the measurements. First, the effective cooling velocity has been determined for each sensor for (1) an idealized probe where the influence of the velocity component tangential to the sensors and flow blockage by the presence of the prongs and the finite lengths of and thermal cross-talk between the sensors are neglected and, (2) for a real probe, the characteristics of which have been determined experimentally. Then, simulating the response of the virtual probes for these two cases to obtain the effective velocities cooling the sensors, velocity and vorticity component statistics have been calculated by assuming the velocity gradients to be constant over the probe sensing area.     

3D reconstruction of the flow and vortical field in a rotating sharp “U” turn channel

Particle image velocimetry experiments have been carried out to obtain visualizations and measurements of the main and secondary flow fields in a square channel with a sharp “U” turn. Both the main and the secondary flow fields have been used to perform a 3D reconstruction of the mean flow and vortical fields in the turn region and in the outlet duct. In order to study the influence of the rotation, tests both in stationary (absence of rotation, Re = 20,000) and in rotating (Re = 20,000 and Ro = 0.3) conditions have been performed. The results show that the Coriolis and centrifugal forces, caused by the rotation, yield strong modifications to the symmetrical flow and vortical fields that are generated, in the static case, only by the abrupt inversion of the flow direction.     

Generation and visualization of volumetric PIV data fields

This review paper addresses the integration of advanced visualization techniques into the analysis of volumetric vector fields obtained by experimental measurement techniques such as holographic PIV, tomographic PIV, 3D PTV or defocusing PIV. The paper follows the idea of the pipeline process for flow visualization focusing on experimental data generation and advanced visualization techniques. The paper tries to help the experimentalist navigating the landscape of recently developed volumetric measurement techniques and advanced visualization techniques. The processing steps and related difficulties are illustrated with the transitional backward facing step flow experiment at Re _h = 4,440. The paper shows the usage of flow visualization for quantitative volumetric PIV data analysis.     

Volumetric reconstruction of the mean flow around circular cylinders fitted with strakes

The volumetric reconstruction technique presented in this paper employs a two-camera stereoscopic particle image velocimetry (SPIV) system in order to reconstruct the mean flow behind a fixed cylinder fitted with helical strakes, which are commonly used to suppress vortex-induced vibrations (VIV). The technique is based on the measurement of velocity fields at equivalent adjacent planes that results in pseudo volumetric fields. The main advantage over proper volumetric techniques is the avoidance of additional equipment and complexity. The averaged velocity fields behind the straked cylinders and the geometrical periodicity of the three-start configuration are used to further simplify the reconstruction process. Two straked cylindrical models with the same pitch (p = 10d) and two different heights (h = 0.1 and 0.2d) are tested. The reconstructed flow shows that the strakes introduce in the wake flow a well-defined wavelength of one-third of the pitch. Measurements of hydrodynamic forces, fluctuating velocity, vortex formation length, and vortex shedding frequency show the interdependence of the wake parameters. The vortex formation length is increased by the strakes, which is an important effect for the suppression of vortex-induced vibrations. The results presented complement previous investigations concerning the effectiveness of strakes as VIV suppressors and provide a basis of comparison to numerical simulations.     

Flow of power-law fluids over a moving wedge surface with wall mass injection

The boundary layer problem of a power-law fluid flow with fluid injection on a wedge whose surface is moving with a constant velocity in the opposite direction to that of the uniform mainstream is analyzed. The free stream velocity, the injection velocity at the surface, moving velocity of the wedge surface, the wedge angle and the power law index of non-Newtonian fluid are assumed variables. The fourth order Runge–Kutta method modified by Gill is used to solve the non-dimensional boundary layer equations for non-Newtonian flow field. Without fluid injection, for every angle of wedge β, a limiting value for velocity ratio λ _cr (velocity of the wedge surface/velocity of the uniform flow) is found for each power-law index n. The value of λ _cr increases with the increasing wedge angle β. The value of wedge angle also restricts the physical characteristics of the fluid to be used. The effects of the different parameters on velocity profile and on skin friction are studied and the drag reduction is discussed. In case of C = 2.5 and velocity ratio λ = 0.2 for wedge angle β = 0.5 with the fluid with power law-index n = 0.5, 48.8% drag reduction is obtained.