Polymer Induced Drag Reduction in a Turbulent Pipe Flow Subjected to a Coriolis Force

The skin friction factor f in a turbulent wall-bounded flow can be greatly reduced by using polymer solutions. In this paper we discuss experimental results on the effect of the Coriolis force on turbulent drag reduction. To study this, a horizontal smooth-walled pipe with internal diameter 25?mm is placed on a horizontal table rotating about its vertical axis. The rotation is made non-dimensional with friction velocity and pipe diameter, to form the Rotation number Ro. For a range of bulk Rotation number (Ro _b ) between 0 and 0.6 for two different Reynolds numbers (Re _b = 15 & 30 × 10³), the pressure drop is measured, from which the average friction factor f is obtained. Additionally the effect of four different polymer concentrations has been investigated. The single-phase results show that the friction factor increases monotonic but gradual with Rotation. With polymer additives a drag reduction is found that increases with concentration, but which is not affected by the rotation.     

Finite element study of free surface turbulent flow in rotating channels

Under certain conditions of liquid flow through rotating channels, the Coriolis force can induce a free surface to be formed. This problem is of practical importance in a Coriolis wear tester, which is used for determining the sliding wear coefficient of wear materials in slurry handling equipment. A deforming Galerkin finite element method is presented for predicting two‐dimensional turbulent free surface mean flow in rotating channels. Reynolds‐averaged Navier–Stokes (RANS) equations are cast into weak(algebraic) form using primitive variables (velocity and pressure). Eddy viscosity is determined via a mixing length model. Velocity is interpolated biquadratically, while pressure is interpolated bilinearly. The kinematic condition is used to form the Galerkin residual for the free surface. The free surface is represented by Hermite polynomials of zeroeth order for continuity of position and slope. Combined Newton's iteration is used to simultaneously solve for the free surface and the field variables. Results of velocity and pressure fields, as well as the free surface are shown to converge with mesh‐size refinement. There is excellent respect for mass conservation. Results are presented for various values of Rossby number (Ro) and height‐based Reynolds number (Re_H). Parameter continuation in Ro and Re_H space is used to compute solutions at higher values of flow rate and angular velocity. Copyright © 2002 John Wiley & Sons, Ltd.     

Non-Darcy mixed convection from a vertical plate in saturated porous media-variable surface heat flux

Nonsimilarity solutions for non-Darcy mixed convection from a vertical impermeable surface embedded in a saturated porous medium are presented for variable surface heat flux (VHF) of the power-law form. The entire mixed convection region is divided into two regimes. One region covers the forced convection dominated regime and the other one covers the natural convection dominated regime. The governing equations are first transformed into a dimensionless form by the nonsimilar transformation and then solved by a finite-difference scheme. Computations are based on Keller Box method and a tolerance of iteration of 10⁻⁵ as a criterion for convergence. Three physical aspects are introduced. One measures the strength of mixed convection where the dimensionless parameter Ra^* _x /Pe^3/2 _x characterizes the effect of buoyancy forces on the forced convection; while the parameter Pe _x /Ra^*2/3 _x characterizes the effect of forced flow on the natural convection. The second aspect represents the effect of the inertial resistance where the parameter K′U _∞/ν is found to characterize the effect of inertial force in the forced convection dominated regime, while the parameter (K′U _∞/ν)(Ra^*2/3 _x /Pe _x ) characterizes the effect of inertial force in the natural convection dominated regime. The third aspect is the effect of the heating condition at the wall on the mixed convection, which is presented by m, the power index of the power-law form heating condition. Numerical results for both heating conditions are carried out. Distributions of dimensionless temperature and velocity profiles for both Darcy and non-Darcy models are presented. Received on 26 May 1997     

Thermo-fluid-dynamic analysis of the flow in a rotating channel with a sharp “U” turn

Infrared thermography has been employed to carry out a detailed convective heat transfer measurements at Re?=?20,000 in a two-pass square channel both for the static case (absence of channel rotation) and for the rotating case (Ro?=?0.3). At the same time, the main and secondary flow fields have been measured by means of particle image velocimetry with the aim to investigate how the flow behavior affects the local distributions of the convective heat transfer coefficient for the two cases. The normal-to-wall velocity component (w) and the turbulent kinetic energy, both measured close to the heat exchanging wall, have been used to formulate an empirical heat transfer correlation within an attempt to identify the role performed by these two quantities on the convective heat transfer coefficient distributions. The latter ones have been reported in terms of normalized Nusselt number (Nu/Nu*) maps, where Nu* is the Nusselt number evaluated with the classical Dittus-B?lter correlation.     

Effect of system rotating on turbulent boundary layer flow

This paper experimentally investigated the effect of rotating on the turbulent boundary layer flow using hot-wire. The experiments were completed in a rotating rig with a vertical axis and four measured positions along the streamwise direction in channel, which focuses on the flow flied in the rotating channel. The rotating effects on velocity profile, wall shear stress and semi-logarithmic mean velocity profile are discussed in this paper. The results indicated that: due to the Coriolis force induced by rotating, the phenomenon of velocity deficit happens near the leading side. The velocity deficit near the leading side, do not increase monotonically with the increase of Ro. The trend of the velocity deficit near the leading side is also affected by the normal component of pressure gradient, which is another important force in the cross-section of the rotating channel. The wall shear stress near the trailing side is larger than that on the leading side, and the semi-logarithmic mean velocity profile is also different under rotating effects. The phenomenon reveals that the effect of rotation penetrates into the logarithm region, and the flow near the leading side tends to turn into laminar under the effect of rotation. The rotation correction of logarithmic law is performed in current work, which can be used in the wall function of CFD to increase the simulating accuracy at rotating conditions.     

Inertial and coriolis effects on oscillatory flow in a horizontal dendrite layer

Flow instability due to oscillatory modes of disturbances in a horizontal dendrite layer during alloy solidification is investigated under an external constraint of rotation. The flow in the dendrite layer, which is modeled as flow in a porous layer and with the inertial effects included, is assumed to rotate about the vertical axis at a constant angular velocity. The investigation is an extension of the work in Riahi (On stationary and oscillatory modes of flow instablity in a rotating porous layer during alloy solidification. J. Porous Media, 6, 177–187, 2003), which was for the case in the absence of the inertial effects. Results of the stability analyses indicate, in particular, that the Coriolis effect can enhance the physical domain for the oscillatory flow, while the inertial effect tends to reduce such domain. Sufficiently strong inertial effect can eliminate presence of the oscillatory mode only for the rotation rate beyond some value. The effect of interaction between the local volume fraction of solid and the flow associated with the Coriolis term was found to be stabilizing.     

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.     

A local lagrangian analysis of passive particle advection in a gas flow field

A local analysis is performed to study the departure from passive advection by small inertial particles based on a Lagrangian framework. The analysis considers heavy particles immersed in a gaseous flow and is restricted to short times, making it relevant to the PIV technique. A necessary (but not sufficient condition) for passive particle advection of inertial particles is that the quantity Λ_maxτ_p be much smaller than one, where Λ_max is the largest modulus of the eigenvalues corresponding to the velocity gradient tensor. This allows for the inertial and passive time scales to match beyond the initial transient, and consequently for the respective trajectories to remain relatively close. Due to this important role regarding advection behavior, Λ_maxτ_p is offered as a definition of a local Stokes number, St_Λ. Since this quantity is a field quantity, it directly provides indication of when and where passive advection by particles can be expected. A departure equation is obtained in one-dimension, where the influence of initial velocity and gravity are explicitly shown. If the flow is irrotational, the higher dimensional analysis reduces to a set of decoupled one-dimensional equations acting along each respective eigenvector of the velocity gradient tensor. A similar expression is found for the case of a purely temporal flow field.     

Intensity of convective mass/heat transfer in a rotary regenerator rotor with the transverse needle-fins

A forced convective mass transfer coefficient was electrochemically measured for a cylindrical bundle of transverse needle-fins ?1 × 10.9, applied as the rotor porous matrix of a rotary heat regenerator. The baffle inside the rotor was present. The technique based on the ferricyanide–ferrocyanide redox reaction controlled at the cathode, in the presence of a sodium hydroxide based electrolyte, was used in this experiment. A set of the six neighbouring fins, connected in parallel, was the cathode. The distribution of the mass transfer coefficient according to different static rotor angle position and the mean mass transfer Chilton–Colburn coefficient correlation j _M  = j _M (Re) for rotation numbers, Ro: 0, 0.8, 1.6 and 2.0 were stated in the mean Reynolds number, Re, range 180–985. The comparison was made between the convective heat fluxes of the pin-fins and the sheet rotor, for Ro = 0.     

The Use of Porous Fins for Heat Transfer Augmentation in Parallel-Plate Channels

In this study, a steady, fully developed laminar forced convection heat augmentation via porous fins in isothermal parallel-plate duct is numerically investigated. High-thermal conductivity porous fins are attached to the inner walls of two parallel-plate channels to enhance the heat transfer characteristics of the flow under consideration. The Darcy–Brinkman–Forchheimer model is used to model the flow inside the porous fins. This study reports the effect of several operating parameters on the flow hydrodynamics and thermal characteristics. This study demonstrates, mainly, the effects of porous fin thickness, Darcy number, thermal conductivity ratio, Reynolds number, and microscopic inertial coefficient on the thermal performance of the present flow. It is found that the highest Nusselt number is achieved at fully filled porous duct which requires the highest pumping pressure. The results show that using porous fins requires less pumping pressure with comparable high heat augmentation weight against fully filled porous duct. It is found that higher Nusselt numbers are achieved by increasing the microscopic inertial coefficient (A), the Reynolds number (Re), and the thermal conductivity of the porous substrate k ₂. The results show that heat transfer can be enhanced (1) with the use of high thermal conductivity fins, (2) by decreasing the Darcy number, and (3) by increasing microscopic inertial coefficient.     

Numerical and experimental investigation of thermal convection for a thermodependent Herschel-Bulkley fluid in an annular duct with rotating inner cylinder

Thermal convection for an incompressible Herschel-Bulkley fluid along an annular duct, whose inner cylinder is rotating and outer is at rest, is analyzed numerically and experimentally. The outer cylinder is heated at constant heat flux density and the inner one is assumed adiabatic. The first part of this study deals with the effect of the rheological behavior of the fluid and that of the rotation of the inner cylinder on the flow field and heat transfer coefficient. All the physical properties are assumed constant and the flow is assumed fully developed. The critical Rossby number Ro_c = (R₁Ω/U_d)_c, for which the dimension of the plug flow is reduced to zero is determined with respect to the flow behavior index, the radius ratio and the Herschel-Bulkley number for axial flow. The rotation of the inner cylinder induces a decrease of the axial velocity gradient at the outer cylinder thereby reducing the heat transfer between the heated wall and the fluid. The second part of this study introduces the variation of the consistency K with temperature and analyzes the evolution of the flow pattern and heat transfer coefficient along the heating zone. Two cases are distinguished depending on the Rossby number: (i) Ro < Ro_c, the plug flow dimension increases along the heating zone; (ii) Ro < Ro_c, the decrease of K with temperature leads to the reappearance of the plug flow. For high angular velocities, it is possible to have a plug zone attached to the outer cylinder. Finally, a correlation is proposed for the Nusselt number. It shows clearly that the effect of thermodependency of K on the heat transfer becomes more important with increasing rotational velocity of the inner cylinder.     

Prediction of unobstructed flow for co‐rotating multi‐disk drive in an enclosure

The flow structure and isotherms of hard disk drives (HDD) are investigated using a finite element method (FEM). The governing equations are based on the three‐dimensional axisymmetric Navier–Stokes partial differential equations (PDEs) with Galerkin FE formulation. Co‐rotating models are selected that include the non‐ventilated configuration within an enclosure. With various operating conditions for the disk system, the following important parameters are considered: disk number (n), rotational speed (Ro), and wall temperature. The flow structure changes rapidly when the rotational Reynolds number (Re_ϕ) is increased. The flow has a greater tendency to flow radially outwards and the swirling velocity tends to be more vertically orientated, especially for high Re_ϕ values. The isotherms only have small varying regions near the rotating axis, forming outward arcs near the wall and inward arcs near the end gap of the disk. Different from the case without the enclosure, the vorticities exist along the outer disk ends. Both the swirling velocity and isotherms indicate nearly symmetrical characteristics, as expected. A higher temperature gradient occurs near the right outer disk ends, which implies the characteristic of higher heat flux. A commercial computational fluid dynamic (CFD) code, CFX‐5, was chosen to validate the results. Copyright © 2001 John Wiley & Sons, Ltd.     

Free convection heat transfer from an isothermal wavy surface in a porous enclosure

The coupled streamfuction–temperature equations governing the Darcian flow and convection process in a fluid-saturated porous enclosure with an isothermal sinusoidal bottom sun face, has been numerically analyzed using a finite element method (FEM). No restrictions have been imposed on the geometrical non-linearity arising from the parameters like wave amplitude (a), number of waves per unit length (N), wave phase (Φ), aspect ratio (A) and also on the flow driving parameter Rayleigh number (Ra). The numerical simulations for varying values of Ra bring about interesting flow features, like the transformation of a unicellular flow to a multicellular flow. Both with increasing amplitude and increasing number of waves per unit length, owing to the shift in the separation and reattachment points, a row–column pattern of multicellular flow transforms to a simple row of multicellular flow. A cycle of n celluar and n+1 cellular flows, with the flow in adjacent cells in the opposite direction, periodically manifest with phase varying between 0 and 360°. The global heat transfer into the system has been found to decrease with increasing amplitude and increasing number of waves per unit length. Only marginal changes in the global heat flux are observed, either with increasing Ra or varying Φ. Effectively, sinusoidal bottom surface undulations of the isothermal wall of a porous enclosure reduces the heat transfer into the system. © 1998 John Wiley & Sons, Ltd.     

Numerical analysis of the rotating viscous flow approaching a solid sphere

A numerical simulation is performed to investigate the flow induced by a sphere moving along the axis of a rotating cylindrical container filled with the viscous fluid. Three‐dimensional incompressible Navier–Stokes equations are solved using a finite element method. The objective of this study is to examine the feature of waves generated by the Coriolis force at moderate Rossby numbers and that to what extent the Taylor–Proudman theorem is valid for the viscous rotating flow at small Rossby number and large Reynolds number. Calculations have been undertaken at the Rossby numbers (R_o) of 1 and 0.02 and the Reynolds numbers (R_e) of 200 and 500. When R_o=O(1), inertia waves are exhibited in the rotating flow past a sphere. The effects of the Reynolds number and the ratio of the radius of the sphere and that of the rotating cylinder on the flow structure are examined. When R_o ? 1, as predicted by the Taylor–Proudman theorem for inviscid flow, the so‐called ‘Taylor column’ is also generated in the viscous fluid flow after an evolutionary course of vortical flow structures. The initial evolution and final formation of the ‘Taylor column’ are exhibited. According to the present calculation, it has been verified that major theoretical statement about the rotating flow of the inviscid fluid may still approximately predict the rotating flow structure of the viscous fluid in a certain regime of the Reynolds number. Copyright © 2004 John Wiley & Sons, Ltd.     

Über eine Verallgemeinerung des Korrekturfaktors K=(Prb/Prw)P,der den Einfluß temperaturabhängiger Stoffwerte auf den konvektiven Wärmeübergang berücksichtigt

The correction factorK, defined byk=(Nu _bH 3.66)/ (Nu _bo 3.66), is applicable only for the purely convective part of heat transfer in tubes. The value 3.66 is the contribution of molecular transport. A widely accepted correlation for the correction factor isK=(Pr_b/Pr_w)^p withp=0.11 for heating and cooling with laminar and turbulent fluid flow. This correlation is not in agreement with theory of similarity. Based on a few restrictive assumptions the author proves, that the exponentp is not a constant but is a function of the following dimensionless parameters: L/D;Pe; ?_Pr andRe. Without the introduction of restrictive assumptions the correction factor is at least a function of 11 dimensionless parameters for some important technical applications. Basically it is conceivable to solve the appropriate differential equations for the considered heat transfer conditions by rather complex numerical methods only for laminar flow.     

Numerical investigation of flow and combined natural-forced convection from an isothermal square cylinder in cross flow

Unsteady flow and heat transfer from a horizontal isothermal square cylinder is studied numerically using a three-dimensional computational model to investigate the influence of buoyancy on the forced flow and heat transfer characteristics. The numerical model is based on a horizontal square cylinder subjected to laminar fluid flow in an unconfined channel. The governing equations in 3D form are solved using a fractional step method based on the finite difference discretization in addition to a Crank–Nicholson scheme employed to the convective and the viscous terms. Two working fluids–air (Pr = 0.7) and water (Pr = 7)–are considered, and the flow and heat transfer simulations were carried out for the Reynolds and Richardson numbers in the intervals 55 ≤ Re ≤ 250 and 0 ≤ Ri ≤ 2, respectively. The flow characteristics such as time-averaged drag/lift, rms drag/rms lift coefficients as well as Strouhal number were computed. The heat transfer from the cylinder is assessed by mean Nusselt number (and rms Nusselt number) over the total heated cylinder walls. As the buoyancy increases, the mass and the velocity of the fluid flowing underneath the cylinder increases. The fluid is injected into the near wake region with an upward motion which significantly alters the flow field in the downstream as well as upstream regions. The effects of Reynolds, Richardson and Prandtl numbers on the flow field and temperature distributions are discussed in detail. It is shown that the flow and heat transfer characteristics are influenced more for air than water. To fill the void in the literature, useful empirical correlations of practical importance are derived for pure forced and pure natural as well as mixed convection. The mixed convection correlations, in terms of the ratio of pure forced convection, are also developed, and their implications are discussed.     

Motion in a layer of rotating liquid,excited by pressure applied to the free surface

It is well known that, in a layer of an ideal incompressible liquid, only surface gravitational waves can exist [1, 2]. If the liquid is rotating, the existence of inertial waves then becomes possible. There have recently appeared a number of communications devoted to the theoretical [3–7] and experimental [8] study of inertial waves. The present article, in the linear statement, discusses the problem of not-fully-established motions, arising in a rotating layer of a homogeneous incompressible liquid under the action of atmospheric pressure applied to the free surface. It is shown that surface and inertial waves arise, and that a geostrophic flow is formed. Asymptotic formulas are written and analyzed for surface and inertial waves. For actual cases of the assignment of the atmospheric pressure, the geostrophic flow is calculated exactly.     

Fully Developed Mixed Convection Flow Between Inclined Parallel Plates Filled with a Porous Medium

A fully developed mixed convection flow between inclined parallel flat plates filled with a porous medium is considered through which there is a constant flow rate and with heat being supplied to the fluid by the same uniform heat flux on each plate. The equations governing this flow are made non-dimensional and are seen to depend on two dimensionless parameters, a mixed convection parameter λ and the Péclet number Pe, as well as the inclination γ of the plates to the horizontal. The velocity and temperature profiles are obtained in terms of λ, Pe and γ when the channel is inclined in an upwards direction as well as for horizontal channels. The limiting cases of small and large λ and small Pe are considered with boundary-layer structures being seen to develop on the plates for large values of λ.     

Large Scale Accumulation Patterns of Inertial Particles in Wall-Bounded Turbulent Flow

Turbulent internal flow in channel and pipe geometry with a diluted second phase of inertial particles is studied numerically. Direct numerical simulations (DNS) are performed at moderate Reynolds number (Re _?? ????200) in pipe and two channels??a smaller one similar in size to previous studies and a 3?×?3-times larger one??and Eulerian statistics pertaining to the particle concentration are evaluated. This simulation box constitutes the largest domain used for particle-laden flows so far. The resulting two-point correlations of the particle concentration show that in the smaller channel the particles organize in thin, streamwise elongated patterns which are very regular and long. The spanwise spacing of these structures is 120 and 160 plus units for the channel and pipe, respectively. Only in the larger box, the streamwise extent is long enough for the particle streaks to decorrelate, thus allowing the particles to move more freely. The influence of the box size on the characteristics of the turbophoresis is clearly shown; a 10% increase of the near-wall correlation is observed for particles with Stokes number St ^?+??=?50. It is thus shown that the box dimensions are an important factor in correctly assessing the motion of inertial particles, and their relation to the underlying velocity field. In addition the binning size effects on the correlation statistics of particle concentration are exploited. In particular the spanwise correlation peak values appear very sensitive to the adopted binning size, although the position of these peaks is found almost independent. Hence to allow a significant comparison between data of different configurations it is necessary to adopt the same binning spacing in inner variable.