Spiral and Wavy Vortex Flows in Short Counter-Rotating Taylor–Couette Cells

Differentially rotating cylinders result in a rich variety of vortical flows for cylindrical Couette flow. In this study we investigate the case of a short, finite-length cavity with counter-rotating cylinders via direct numerical simulation using a three-dimensional spectral method. We consider aspect ratios ranging from 5 to 6. Two complex flow regimes, wavy vortices and interpenetrating spirals, occur with similar appearance to those found experimentally for much larger aspect ratios. For wavy vortices the wave speed is similar to that found for counter-rotating systems and systems in which the outer cylinder is stationary. For the interpenetrating spiral structure, the vortices are largely confined to the unstable region near the inner cylinder. The endwalls appear to damp and stabilize the flow as the aspect ratio is reduced to the point that in some cases the vortical flow is suppressed. At higher inner cylinder speeds, the interpenetrating spirals acquire a waviness and the vortices, while generally near the inner cylinder, can extend all of the way to the outer cylinder. Received 5 November 2001 and accepted 29 March 2002 Published online 2 October 2002 Communicated by H.J.S. Fernando     

Flow in a cylinder driven by rotating split-disk endwalls

Flow of an incompressible viscous fluid contained in a cylindrical vessel (radius R, height H) is considered. Each of the cylinder endwalls is split into two parts which rotate steadily about the central axis with different rotation rates: the inner disk (r < r₁) rotating at Ω₁, and the outer annulus (r₁ < r < R) rotating at Ω₂. Numerical solutions to the axisymmetric Navier-Stokes equations are secured for small system Ekman numbers E ( v/(ΩH²)). In the linear regime, when the Rossby number Ro

Full-size image (1K)

, the numerical results are shown to be compatible with the theoretical prediction as well as the available experimental measurements. Emphasis is placed on the results in the nonlinear regime in which Ro is finite. Details of the structures of azimuthai and meridional flows are presented by the numerical results. For a fixed Ekman number, the gross features of the flow remain qualitatively unchanged as Ro increases. The meridional flows are characterized by two circulation cells. The shear layer is a region of intense axial flow toward the endwall and of vanishing radial velocity. The thicknesses of the shear layer near r = r₁ and the Ekman layer on the endwall scale with E and E , respectively. The numerical results are consistent with these scalings.     

Numerical simulations of turbulent shear flows

We simulate three-dimensional mixing layers, isotropic turbulence, and rotating turbulence. In the mixing-layer case, we show that high compressibility inhibits helical pairings obtained in the incompressible case, yielding a staggered array of large -shaped vortices. For isotropic turbulence, one shows the existence of large coherent low-pressure vortices, characterized by exponential tails of various p.d.f.'s. We develop also a new subgrid-scale model applied with success to the backward-facing step, and to the compressible boundary layer. Finally, one looks at the influence of solid-body rotation upon free-shear flows or homogeneous turbulence. At moderate Rossby numbers, cyclonic vortices are stabilized, while anticyclonic vortices are disrupted into intense Görtler-type alternate longitudinal vortices.     

Development of a local disturbance in the boundary layer on a curved surface

The process of nonlinear development of a local transverse disturbance on a concave surface is analyzed and the mechanism of formation of the resulting periodic structure is examined. Attention is concentrated on a qualitative analysis of the flow. Equations describing the development of a transverse disturbance in a laminar boundary layer are obtained on the basis of the asymptotic behavior of the Navier-Stokes equations as Re . A solution describing the Taylor vortices formed between two coaxial cylinders when the inner cylinder rotates is obtained. The experimental data on Görtler vortices in boundary layers are analyzed.Translated from Izvestiya Akademii Nauk SSSR, Mekhanika Zhidkosti i Gaza, No. 1, pp. 23–31, January–February, 1990.     

An experimental study on vortex breakdown in a differentially-rotating cylindrical container

The vortex breakdown phenomenon in a closed cylindrical container with a rotating endwall disk was reproduced. Visualizations were performed to capture the prominent flow characteristics. The locations of the stagnation points of breakdown bubbles and the attendant global flow features were in excellent agreement with the preceding observations. Experiments were also carried out in a differentially-rotating cylindrical container in which the top endwall rotates at a relatively high angular velocity _t, and the bottom endwall and the sidewall rotate at a low angular velocity _sb. For a fixed cylinder aspect ratio, and for a given relative rotational Reynolds number based on the angular velocity difference _t–_sb, the flow behavior is examined as |_sb/_t| increases. For a co-rotation (_sb/_t>0), the breakdown bubble is located closer to the bottom endwall disk. However, for a counter-rotation (_sb/_t<0), the bubble is seen closer to the top endwall disk. For sufficiently large values of _sb, the bubble ceases to exist for both cases.     

Wind Impact on Single Vortices and Counterrotating Vortex Pairs in Ground Proximity

Wall-resolved large eddy simulations are employed to investigate the behaviour of wake vortices and single vortices in ground proximity at a variety of wind conditions. The six considered strengths of wind, ranging between 0.5 and 4 times the initial wake vortex descent speed, w₀, include practically and theoretically significant wind speeds. A crosswind of 0.5 w₀ may lead to windward stall posing a potential hazard to subsequently landing aircraft, whereas theoretical considerations predict that at 4 w₀ the rebound of the luff vortex is completely suppressed. The same range of wind speeds is also used to investigate the effects of headwind and diagonal wind in order to discriminate between effects of environmental turbulence increasing with wind speed and the direction of the wind shear. The study has been complemented by a number of single vortex computations in order to differentiate between effects related to the mutual interaction of the vortex pair and the individual vortices with the turbulent boundary layer flow. It is shown that vortex ascent, descent, rebound and decay characteristics are controlled by (i) the interaction of the vortices with secondary vorticity detaching from the ground, (ii) the redistribution of vorticity of the boundary layer which is altering the path of the primary vortices by mutual velocity induction, and (iii) the interaction of the vortices with the environmental turbulence.     

Strong vortex/boundary layer interactions

Detailed measurements with hot-wires and pressure probes are presented for the interaction between a turbulent longitudinal vortex pair with common flow down, and a turbulent boundary layer. The interaction has a larger value of the vortex circulation parameter, and therefore better represents many aircraft/vortex interactions, than those studied previously. The vortices move down towards the boundary layer, but only the outer parts of the vortices actually enter the it. Beneath the vortices the boundary layer is thinned by lateral divergence to the extent that it almost ceases to grow. Outboard of the vortices the boundary layer is thickened by lateral convergence. The changes in turbulence structure parameters in the boundary layer appear to be due to the effects of extra-rate-of-strain produced by lateral divergence (or convergence) and by free-stream turbulence. The effect of the interaction on the vortices (other than the inviscid effect of the image vortices below the surface) is small. The flow constitutes a searching test case for prediction methods for three-dimensional turbulent flows.     

Laminar boundary layer on blunt bodies with account for vorticity of the external flow

This paper presents the technique for and results from numerical calculations of the hypersonic laminar boundary layer on blunted cones with account for the vorticity of the external flow caused by the curved bow shock wave. It is assumed that the air in the boundary layer is in the equilibrium dissociated state, but the Prandtl number is assumed constant, =0.72. The calculations were made in the range of velocities 3–8 km/sec, cone half-angles _k=0°–20°. With account for the vortical interaction of the boundary layer with the external flow, the distribution of the thermal flux and friction will depend on the freestream Reynolds number (other conditions being the same). In the calculations the Reynolds number R, calculated from the freestream parameters and the radius of the spherical blunting, varies from 2.5·10³ to 5.10⁴. For the smaller Reynolds numbers the boundary layer thickness on the blunting becomes comparable with the shock standoff, and for R<2.5·10³ it is apparent that we must reconsider the calculation scheme. With R>5·10⁴ for cones which are not very long the vortical interaction becomes relatively unimportant. The results of the calculations are processed in accordance with the similarity criteria for hypersonic viscous gas flow past slender blunted cones [1, 2].     

Numerical investigation of secondary flows in a high-lift low pressure turbine

In turbomachines, secondary flows (or endwall flows) typically originate at the junction between endwalls and the blade surface. Within the blade passage, the strength of the secondary flows is amplified by the crossflow from the pressure to the suction surface of the blade. The enhanced mixing due to secondary flows induce additional losses into the system. This decreases the overall work output and also changes the flow incidence onto the downstream blade rows. Using a series of high-fidelity eddy resolving simulations, the current study attempts to provide an improved understanding for the complex flow physics over the endwalls of a high-lift Low Pressure Turbine (LPT) blade. The effect of three different inflow conditions has been studied. These include a laminar boundary layer (LBL), a turbulent boundary layer (TBL) and wakes with secondary flow (W&S) from an upstream blade row. For the simulations with TBL and W&S, precursor eddy resolving simulations were used to prescribe realistic inflows. The loss generation mechanisms were subsequently studied both at the endwall and the midspan, which includes evaluating the mass-averaged total pressure loss coefficient (Y_p) and the loss generation rate.When compared to LBL, additional disturbances from an incoming TBL and wakes with secondary flows enhanced the mixing within the blade passage resulting in a substantial increase in the total pressure loss. Prior to flow transition, incoming wakes with secondary flows increased the local loss generation rate at both the endwall and the midspan in the front portion of the blade passage (x/C_x < 0.84). In contrast, in the aft portion of the passage (x/C_x > 0.8), the incoming wakes effectively suppressed the separation bubble at the midspan thereby decreasing the local loss generation rate. It is also demonstrated that the wakes shed from the trailing edge at the mid-span mix out rapidly when compared to the passage vortex at the endwall.     

The shape of a liquid surface between rotating concentric cylinders

Summary The shape of the free surface of a second order fluid, enclosed between vertical concentric cylinders with the inner cylinder rotating with angular velocity, is calculated. The calculation uses a perturbation procedure in which, for slow flows, the boundary conditions on the free surface are transformed to boundary conditions on the fixed horizontal surface obtained when is zero. The results are tested experimentally.With 3 figures     

Two-scale semiempirical theory of turbulent boundary layers and jets

To characterize the turbulence of boundary layers in the energy-bearing interval of wave numbers several turbulence scales are sometimes used (for example, [1, 2]). In particular, the universality of the semiempirical model of turbulence [2] can be extended in this way. A turbulence model with one equation (energy balance of the turbulence) has been constructed and used [3–6] and it has been established that the number of problems that can be solved for a universal choice of the values of the empirical coefficients increases appreciably if not one but two turbulent scales are used. In the present paper, it is shown that the introduction of a second scale makes it possible to take into account the interaction of shear layers in flows with two shear layers (for example, a channel or jet), and also to take into account the influence of turbulence of an external flow on a boundary layer. The interaction of shear layers is taken into account in theories containing a transport equation for the turbulent frictional stress _t (for example, [7]), in which the essence of the interaction reduces to diffusion of _t from layer to layer. In the present paper, a predominant volume interaction effect is assumed. It takes the form of a difference between the interaction of large-scale vortices with a shear deformation motion in flows with one and two shear layers, and also in the presence of turbulence in an external flow.Translated from Izvestiya Akademii Nauk SSSR, Mekhanika Zhidkosti i Gaza, No. 8, pp. 17–25, November–December, 1982.     

Wall proximity effects on the effectiveness of upstream control rod

Two dimensional flow over a circular cylinder with an upstream control rod of same diameter is simulated in unbound condition and in wall bounded conditions. The cylinders are placed at various heights from the wall and the inter-distance between cylinders is also varied. The control rod is subjected to different rotation rates. It is found that, in unbound condition, rotating the control rod decreases the critical pitch length (S/D_cr) and increases the drag and Strouhal number of the main cylinder. In presence of plane wall, the shielding provided by the separated shear layers from the control rod in cavity regime is deteriorated due to deflection of shear layers which results in higher drag and large fluctuation of lift coefficient. However, in wake impingement regime, the binary vortices from the control rod are weakened due to diffusion of vorticity and hence, the main cylinder experiences a lower drag and small lift fluctuations than that of unbound condition. The critical height of vortex suppression (H/D_cr) is higher in cavity regime than that of wake impingement regime due to the single extended-bluff body like configuration. The rotation of control rod energizes the wall boundary layer and increases the critical height of vortex suppression. Increasing the rotational rate of control rod decreases the drag force and reduces the amplitude of lift fluctuation. Analysis of the wall shear stress distribution reveals that it suffers a sudden drop at moderate height where the normal Karman vortex shedding changes to irregular shedding consisting of single row of negative vortices. Modal structures obtained from dynamic mode decomposition (DMD) reveal that the flow structures behind the main cylinder are suppressed due to wall and the flow is dominated by the wake of control rod.     

Optical investigation of the heat transfer from a rotating cylinder in a cross flow

The heat transfer from a rotating cylinder in an air-cross flow was investigated by purely optical measuring techniques. Flow velocities were measured by a two-dimensional LDV both in the vicinity of the cylinder and in the boundary layer. A new optical device based on light-deflection in a temperature field was developed to examine local temperature gradients in the boundary layer of the rotating cylinder. Finally, a Michelson-interferometer was installed to produce real-time pictures of isothermal lines around the heated cylinder. The impact of rotation on flow patterns, boundary layer behaviour and heat transfer could be clearly identified. It appears that the velocity-ratio acts like an independent parameter, in that flow patterns correspond to this dimensionless number. Furthermore, it seems that rotation dominates over cross flow, both fluid-dynamically and thermally above = 2.This work was carried out at the University of the Federal Armed Forces in Munich/Germany.     

Nonlinear dynamics of large-scale vortex structures in a turbulent Ekman layer

Nonlinear regimes of roll structure development in the Ekman atmospheric boundary layer are considered. Within the framework of a turbulent boundary layer theory that takes the inherent helicity into account, the helicity effect on the stability and development of secondary vortices is studied. The stabilizing role of helicity in the Ekman layer dynamics is detected. With increasing nonlinearity the rolls become time-dependent self-oscillatory structures exchanging energy and helicity with the main flow. Multi-scale structure regimes are obtained, similar to those observable in the atmospheric boundary layer and the solitary vortex structures previously discovered in bench and numerical experiments for Ekman flows between disks rotating in opposite directions.     

PIV analysis of near-wake behind a sphere at a subcritical Reynolds number

The vortical structure of near-wake behind a sphere is investigated using a PIV technique in a circulating water channel at Re = 11,000. The measured velocity fields show a detailed vortical structure in the recirculation region such as recirculation vortices, reversed velocity zone, and out-of-plane vorticity distribution. The vorticity distribution of the sphere wake shows waviness in cross-sectional planes. The time-averaged turbulent structures are consistent with the visualized flow showing the onset of shear layer instability. The spatial distributions of turbulent intensities provide turbulent statistics for validating numerical predictions.     

Theory of gravitational stability of a rotating cylinder

The stability of a rotating dust cylinder against perturbations located in the plane perpendicular to the axis of rotation is investigated. It is shown that a homogeneous rotating cylinder containing a weak inhomogeneity is stable against such perturbations. A weakly inhomogeneous cylinder with opposite streams of equal density is unstable for thel=2 mode in the case of a perturbation of the form_ei(l–t), when the density increases radially. The instability of a system consisting of a homogeneous rotating dust cylinder in a hot homogeneous medium is determined. It is shown that the maximum growth rate corresponds tol = 2 when the density of a cold cylinder is not negligible in comparison with the density of the medium. In the opposite case, the maximum growth rate shifts toward l=3. An attempt is made to associate the existence of the maximum growth rate for l=2 with the presence of two spiral arms in most galaxies. It is shown that, when the longitudinal temperature is high enough, a rotating cylinder which is bounded in the radial direction is stable against arbitrary perturbations.Translated from Zhurnal Prikladnoi Mekhaniki i Tekhnicheskoi Fiziki, Vol.10, No. 3, pp. 3–11, May–June, 1969.     

Three-dimensional gas flow in a rotating cylinder with a retarding cover

The study of rotating flows is of interest due to both the development of the centrifugal method of separation of gas and isotope mixtures and the possibility of astrophysical applications. An analytical nonlinear model for calculating the hydrodynamic characteristics of the viscous incompressible fluid flow in a rotating cylinder in the presence of a retarding cover is presented. The cases of stationary and rotating covers are considered. The analysis is performed on the basis of the system of hydrodynamic Navier-Stokes equations. The flow domain is divided up into the main flow and end boundary layers at the cylinder bottom and at the rotating cover. In its turn, the main flow is divided up into an inviscid quasi-rigid core and a lateral layer within which almost the entire upward circulatory flow is concentrated. The equations of the boundary layers at the end surfaces are analyzed by the approximate Slezkin-Targ method. The solutions in the boundary and lateral layers are “stitched” together with the velocity distribution in the main flow core. The unknown angular velocity ω ₁ and radial boundary R ₁ of the core are determined from the balance of the moments of the friction forces acting on the main rotating flow and the continuity condition for the circulatory flow. The experimental and calculated data are compared.     

Influence of incident vortex street on separated flow around a finite blunt plate: PIV measurement and POD analysis

This study investigated the influence of incident vortex street on the spatial characteristics of separated shear layers around finite blunt plates through use of planar particle image velocimetry. Three systems with different chord-to-thickness ratios (c/t) were chosen for the comparative study (i.e., c/t=3.0, 6.0 and 9.0). The Reynolds number, based on the plates׳ thickness (t), was Re_t=1000. The incident vortex street was generated by placing a circular cylinder (D=t) far upstream of the plate. For the systems without incident vortex street, the separated shear layers around the shortest (c/t=3.0), the median-length plate (c/t=6.0) and the longest (c/t=9.0) plates had no re-attachment, periodic re-attachment and faithful re-attachment on the plate׳s surface, respectively. However, the separated shear layers subjected to incident vortex street were restrained at the leading edges by the upstream vortical structures, which were less influenced by increases in chord-to-thickness ratio. Contour plots of the spatial v–v correlation coefficient revealed that the wakes behind plates longer than c/t=6.0 were not severely influenced by incident vortex street. Distributions of the spatial v–v correlations and the POD eigenmodes revealed that the incident vortical structures were split by the leading edges of the plates in all systems, resulting in two vortices with the same rotating direction. Subsequently, a further phase-averaged analysis convincingly demonstrated the splitting process of the incident vortical structures by the leading edges.     

High Reynolds number thick axisymmetric turbulent boundary layer measurements

Experimental measurements of the wall shear stress and momentum thickness for thick axisymmetric turbulent boundary layers are presented. The use of a full-scale towing tank allowed zero pressure gradient turbulent boundary layers to be developed on cylinders with diameters of 0.61, 0.89, and 2.5 mm and lengths ranging from 30 m to 150 m. Moderate to high Reynolds numbers (10⁴<Re <10⁵, 10⁸<Re _L<10⁹) are considered. The relationship between the mean wall shear stress, cylinder diameter, cylinder length, and speed was investigated, and the spatial growth of the momentum thickness was determined. The wall shear stress is significantly higher, and the spatial growth of the boundary layers is shown to be lower than for a comparable flat-plate case. The mean wall shear stress exhibits variations with length that are not seen in zero pressure gradient flat plate turbulent boundary layers. The ratio of outer to inner boundary layer length scales is found to vary linearly with Re , which is qualitatively similar to a flat plate turbulent boundary layer. The quantitative effect of a riblet cylindrical cross-sectional geometry scaled for drag reduction based on flat plate criteria was also measured. The flat plate criteria do not lead to drag reduction for this class of boundary layer shear flows.List of symbols a cylinder radius, mm - A _s total cylindrical surface area, m² - C _d tangential drag coefficient - D drag force, Newtons - boundary layer thickness, mm - * displacement thickness, mm - h riblet height, mm - L cylinder length, m - kinematic viscosity, m²/s - momentum thickness, mm - fluid density, kg/m³ - r radial coordinate, mm - Re _L Reynolds number based on length= - Re Reynolds number based on momentum thickness= - s riblet spacing, mm - _w mean wall shear stress, N/m² - u(r) mean streamwise velocity, m/s - u friction velocity= - U _o tow speed, m/s - x streamwise coordinate, m