Numerical modeling of laminar-turbulent transition in a boundary layer at a high freestream turbulence level

Disturbances generated by external turbulence in the boundary layer on a flat plate set suddenly in motion are determined by numerically solving the Navier-Stokes equations. The results of direct numerical simulation of isotropic homogenous turbulence are taken as initial conditions. The solution obtained models laminar-turbulent transition in the flat-plate boundary layer at a high freestream turbulence level, time measured from the onset of the motion serving as the longitudinal coordinate. The solution makes it possible to estimate the effect of different factors, such as flow unsteadiness and nonlinearity and the characteristics of the freestream velocity fluctuation spectrum, on laminar-turbulent transition in the boundary layer.     

Statistical description of laminar-turbulent transition in a boundary layer at high freestream turbulence degree

The statistical approach is applied to calculate the intermittence coefficient in boundarylayer laminar-turbulent transition due to external turbulence. It is assumed that turbulent regions in the boundary layer are associated with the appearance of turbulent spots generated by the secondary instability of streaky structures that have reached a threshold amplitude. An universal dependence of the intermittence coefficient in the transition region on the velocity fluctuation amplitude and the Reynolds number is obtained. It describes well the experimental results.     

Reynolds analogy in a dusty laminar boundary layer

The nonisothermal Blasius problem for a gas suspension is considered on the basis of the equations of a quasiequilibrium two-phase laminar boundary layer [1–3]. Approximate analytical expressions are obtained for the friction and heat transfer coefficients and their region of applicability is estimated; the Reynolds analogy between friction and convective heat transfer processes [4] is extended to the case of a dusty quasiequilibrium laminar boundary layer. Translated from Izvestiya Akademii Nauk SSSR, Mekhanika Zhidkosti i Gaza, No. 6, pp. 160–162, November–December, 1986.     

Asymptotic analysis of turbulent boundary layer flow on a flat plate at high Reynolds numbers

The equations of the turbulent boundary layer contain a small parameter — the reciprocal of the Reynolds number, which makes it possible to carry out an asymptotic analysis of the solutions with respect to that small parameter. Such analyses have been the subject of a number of studies [1–5]. In [2, 5] for closing the momentum equation algebraic Prandtl and turbulent viscosity models were used. In [1, 3, 4] the structure of the boundary layer was analyzed in general form without formulating specific closing hypothesis but under additional assumptions concerning the nature of the asymptotic behavior of the limiting solutions in the various regions.Translated from Izvestiya Rossiiskoi Akademii Nauk, Mekhanika Zhidkosti i Gaza, No.4, pp. 106–117, May-June, 1993.     

Turbulent thermal boundary layer on a plate. Reynolds analogy and heat transfer law over the entire range of prandtl numbers

Arational asymptotic theory is proposed,which describes the turbulent dynamic and thermal boundary layer on a flat plate under zero pressure gradient. The fact that the flow depends on a finite number of governing parameters makes it possible to formulate algebraic closure conditions relating the turbulent shear stress and heat flux with the gradients of the averaged velocity and temperature. As a result of constructing an exact asymptotic solution of the boundary layer equations, the known laws of the wall for velocity and temperature, the velocity and temperature defect laws, and the expressions for the skin friction coefficient, Stanton number, and Reynolds analogy factor are obtained. The latter makes it possible to give two new formulations of the temperature defect law, one of which is identical to the velocity defect law and contains neither the Stanton number nor the turbulent Prandtl number, and the second formulation does not contain the skin friction coefficient. The heat transfer law is first obtained in the form of a universal functional relationship between three parameters: the Stanton number, the Reynolds number, and the molecular Prandtl number. The conclusions of the theory agree well with the known experimental data.     

Double large field stereoscopic PIV in a high Reynolds number turbulent boundary layer

An experiment on a flat plate turbulent boundary layer at high Reynolds number has been carried out in the Laboratoire de Mecanique de Lille (LML, UMR CNRS 8107) wind tunnel. This experiment was performed jointly with LEA (UMR CNRS 6609) in Poitiers (France) and Chalmers University of Technology (Sweden), in the frame of the WALLTURB European project. The simultaneous recording of 143 hot wires in one transverse plane and of two perpendicular stereoscopic PIV fields was performed successfully. The first SPIV plane is 1?cm upstream of the hot wire rake and the second is both orthogonal to the first one and to the wall. The first PIV results show a blockage effect which based on both statistical results (i.e. mean, RMS and spatial correlation) and a potential model does not seem to affect the turbulence organization.     

Direct numerical simulation of turbulent heat transfer on the Reynolds analogy over irregular rough surfaces

The effect of rough surface topography on heat and momentum transfer is studied by direct numerical simulations of turbulent heat transfer over uniformly heated three-dimensional irregular rough surfaces, where the effective slope and skewness values are systematically varied while maintaining a fixed root-mean-square roughness. The friction Reynolds number is fixed at 450, and the temperature is treated as a passive scalar with a Prandtl number of unity. Both the skin friction coefficient and Stanton number are enhanced by the wall roughness. However, the Reynolds analogy factor for the rough surface is lower than that for the smooth surface. The semi-analytical expression for the Reynolds analogy factor suggests that the Reynolds analogy factor is related to the skin friction coefficient and the difference between the temperature and velocity roughness functions, and the Reynolds analogy factor for the present rough surfaces is found to be predicted solely based on the equivalent sand-grain roughness. This suggests that the relationship between the Reynolds analogy factor and the equivalent sand-grain roughness is not affected by the effective slope and skewness values. Analysis of the heat and momentum transfer mechanisms based on the spatial- and time-averaged equations suggests that two factors decrease the Reynolds analogy factor. One is the increased effective Prandtl number within the rough surface in which the momentum diffusivity due to the combined effects of turbulence and dispersion is larger than the corresponding thermal diffusivity. The other is the significant increase in the pressure drag force term above the mean roughness height.     

Smooth and rough turbulent boundary layers at high Reynolds number

A 2-D turbulent boundary layer experiment with zero pressure gradient (ZPG) has been carried out over a rough and a smooth surface using two cross hot-wire probes. Wind tunnel speeds of 10 m/s and 20 m/s were set up in order to investigate the effects of the upstream conditions and the Reynolds number on the downstream flow. For a given set of upstream conditions, such as the wind tunnel speed, trip wire size and location, the three components of the velocity field were measured from about 14 m from the inlet of the wind tunnel to 30 m downstream. This experiment is unique because it achieves Reynolds numbers as high as R120,000, for which measurements of the mean velocity are reported. It is shown that by fixing the upstream conditions, the mean deficit profiles collapse with the freestream velocity, , but to different curves depending on the upstream conditions and surface roughness. Moreover, the effects of the upstream conditions, the Reynolds number, and roughness are completely removed from the outer flow when the mean deficit profiles are normalized by the Zagarola/Smits scaling, . Consequently, the true asymptotic profile in the turbulent boundary layer is found in ZPG flow regardless of the range of Reynolds number, surface conditions and initial conditions.     

Modeling of the flow structure and heat transfer in a gas-droplet turbulent boundary layer

A numerical study of dynamics and heat/mass transfer in a gas-droplet turbulent boundary layer on a vertical flat plate is carried out. A large number of factors which affect the heat and mass transfer and the structure of thermal and concentration fields in a turbulent boundary layer is analyzed. It is shown that the increase in droplet concentration results in the intensification of heat transfer, as compared with the single-phase air flow. The comparison of this analysis with experimental data shows a qualitative and quantitative agreement between the calculated and experimental data.     

On the logarithmic-law constants and the turbulent boundary layer at low Reynolds numbers

It is shown that a family of formally derived similarity solutions describe to leading order the outer region of a turbulent boundary layer for all Reynolds numbers for which the layer satisfies the logarithmic law-of-the-wall. The family includes Coles' [1] hypothesis. For consistency with this hypothesis and the logarithmic law-of-the-wall, it is further shown that the constants in the latter form the product κC=2+O(ε), suggesting the logarithmic law of the wall be written $${U \mathord{\left/ {\vphantom {U {U_\tau = \kappa ^{ - 1} }}} \right. \kern-\nulldelimiterspace} {U_\tau = \kappa ^{ - 1} }}\ln \left( {e^2 U_\tau {y \mathord{\left/ {\vphantom {y \nu }} \right. \kern-\nulldelimiterspace} \nu }} \right) + O\left( \in \right).$$ A range of data are reprocessed to determine the skin friction coefficientC _f using κC = 2 and these collapse well when plotted against momentum thickness Reynolds number, Re _θ . It is also shown that the form parameter, Π, in Coles hypothesis is not unique but is determined by history effects peculiar to the boundary layer. Expressions are derived forC _f (Re _θ ) and the shape factorH (Re _θ ); both agree closely with the data and are valid over all Reynolds numbers for which the logarithmic law of the wall is satisfied.     

Large eddy simulation of a developing turbulent boundary layer at a low Reynolds number

A spectral code has been used to simulate a developing turbulent boundary layer at low Reynolds number Re_θ (based on free stream velocity and momentum thickness) between 353 and 576. The starting field was generated by allowing a step change of temperature to diffuse outwards from one wall in a fully developed channel flow. The thermal boundary layer so created was conditionally sampled to convert it into a momentum boundary layer with an irrotational free stream region, a process which is justified by appeal to experiments. This initial field was allowed to develop until the momentum boundary layer thickness δ₉₉₅ had grown to about 1·5 times its original thickness. The results of the simulation have been compared with a wide range of experimental data. The outcome of this comparison is generally very satisfactory; the main trends of the experiments are well reproduced and our simulation supplements and extends the existing sets of experimental data. The simulation also gives pressure statistics which cannot be obtained experimentally. In particular, it gives the contribution of pressure diffusion to the balance equations for the Reynolds stress and indicates the error produced by omitting this term.     

Effect of turbulence of external flow on the turbulent boundary layer at a plate

In [1, 2] turbulence of the external flow was taken into account by specifying the turbulent energy at the external boundary of the boundary layer on integrating the energy-balance equation for the turbulence. In [3] a special correction that allowed the turbulence of the external flow to be taken into account was introduced in determining the mixture path. In [4, 5] the turbulent energy calculated from the energy-balance equation of the turbulence was added to the energy induced by turbulence of the external flow, the energy distribution of the induced turbulence being specified using an empirically selected function. In [6, 7] a method of taking into account the effect of turbulence of the external flow on a layer of mixing and a jet was proposed. In the present work, this method is applied to the boundary layer at a plate.Translated from Izvestiya Akademii Nauk SSSR, Mekhanika Zhidkosti i Gaza, No. 3, pp. 26–31, May–June, 1977.     

Effect of coarse particles on the heat transfer in a particle-laden turbulent boundary layer

The temperature distribution in particle-laden turbulent flow, in a flume, was investigated both by DNS and experimentally. Simulations were performed at Re=171 and Pr=5.4 in order to study the interaction between the particle motion and flow turbulence. Two-way coupling was used to obtain various turbulence statistics, the grid resolution was sufficiently fine to resolve all essential turbulent scales. The effect of particle diameter on momentum, heat transfer and particle deposition was considered. The details of particle-turbulence interaction depend on the particle Stokes number and the particle Reynolds number.

The spatial structures of instantaneous flow and temperature fields were visualized. Low frequency small oscillations of deposited particles were observed. It was found that these small deviations from the initial position, caused strong changes in the instantaneous temperature field near the particle.

The experiments provided details of the temperature field on the heated wall close to the particle. In the front of the particle, a sharp increase in heat transfer coefficient was observed. The experimental results agree well with the computational predictions.     

Investigation of the heat transfer mechanism in supersonic turbulent boundary layers

A method is described for calculating turbulent Prandtl numbers from Mach number and total temperature profiles in supersonic boundary layers. The calculations are based on boundary layer measurements in the Mach number range from 3.5 to 5. The investigations clearly indicate that in addition to accurate profile measurements reliable values of shear stress and heat flux at the wall must exist, in order to be able to calculate the turbulent Prandtl number in the viscous regime of the boundary layer. For flow conditions with and without heat transfer, the derived turbulent Prandtl numbers indicate that the turbulent transport of heat decreases much faster towards the wall than the turbulent transport of momentum. The results of the analysis show that only the unequivocal qualitative result of increasing turbulent Prandtl numbers in the viscous region of the boundary layer, can be expected. The variation of the turbulent Prandtl number can be described successfully using a simple approximation, based on the mixing length concept, and is applied to the calculation of total temperature distribution using the law of the wall for compressible flow.     

Freestream turbulence effect on flow and heat transfer in the flat-plate boundary layer

Flow and heat transfer in the flat-plate boundary layer is numerically investigated using a differential three-equation turbulence model for the initial freestream turbulence intensity ranging from 1.5 to 9%. An increase in the local friction coefficient and the Stanton number obtained in the calculations is in agreement with the most representative experimental data.     

Low Reynolds number axisymmetric turbulent boundary layer on a cylinder

In the present study, an axisymmetric turbulent boundary layer growing on a cylinder is investigated experimentally using hot wire anemometry. The combined effects of transverse curvature as well as low Reynolds number on the mean and turbulent flow quantities are studied. The measurements include the mean velocity, turbulence intensity, skewness and flatness factors in addition to wall shear stress. The results are presented separately for the near wall region and the outer region using dimensionless parameters suitable for each case. They are also compared with the results available in the open literature.The present investigation revealed that the mean velocity in near wall region is similar to other simple turbulent flows (flat plate boundary layer, pipe and channel flows); but it differs in the logarithmic and outer regions. Further, for dimensionless moments of higher orders, such as skewness and flatness factors, the main effects of the low Reynolds number and the transverse curvature are present in the near wall region as well as the outer region.     

LDA measurements in low Reynolds number turbulent boundary layer

LDA measurements of the mean velocity in a low Reynolds number turbulent boundary layer allow a direct estimate of the friction velocity U from the value of /y at the wall. The trend of the Reynolds number dependence of / is similar to the direct numerical simulations of Spalart (1988).