Bulletin of the American Physical Society
67th Annual Meeting of the APS Division of Fluid Dynamics
Volume 59, Number 20
Sunday–Tuesday, November 23–25, 2014; San Francisco, California
Session L26: Thermal Transport in Boundary Layers |
Hide Abstracts |
Chair: Johan Larsson, University of Maryland Room: 2007 |
Monday, November 24, 2014 3:35PM - 3:48PM |
L26.00001: DNS of transcritical turbulent boundary layers at supercritical pressures under abrupt variations in thermodynamic properties Soshi Kawai In this talk, we first propose a numerical strategy that is robust and high-order accurate for enabling to simulate transcritical flows at supercritical pressures under abrupt variations in thermodynamic properties due to the real fluid effects. The method is based on introducing artificial density diffusion in a physically-consistent manner in order to capture the steep variation of thermodynamic properties in transcritical conditions robustly, while solving a pressure evolution equation to achieve pressure equilibrium at the transcritical interfaces. We then discuss the direct numerical simulation (DNS) of transcritical heated turbulent boundary layers on a zero-pressure-gradient flat plate at supercritical pressures. To the best of my knowledge, the present DNS is the first DNS of zero-pressure-gradient flat-plate transcritical turbulent boundary layer. The turbulent kinetic budget indicates that the compressibility effects (especially, pressure-dilatation correlation) are not negligible at the transcritical conditions even if the flow is subsonic. The unique and interesting interactions between the real fluid effects and wall turbulence, and their turbulence statistics, which have never been seen in the ideal-fluid turbulent boundary layers, are also discussed. [Preview Abstract] |
Monday, November 24, 2014 3:48PM - 4:01PM |
L26.00002: Design and Validation of a Constant Wall Temperature Plate Drummond Biles, Alireza Ebadi, Allen Ma, Chris White A thermally conductive constant temperature wall-plate has been constructed and wind tunnel validation tests of the wall-plate design have been performed. The wall-plate is a sectioned wall design, where each section is independently heated and controlled. Each section consists of an aluminum 6061 plate, an array of resistive heaters affixed to the bottom of the aluminum plate, and a calcium silicate holder used for thermal isolation. A 3 $\times$ 3 grid of embedded thermocouples in each aluminum plate are used to monitor wall temperature and for feedback control of wall heating. The streamwise (flow direction) length of each section increases with downstream position since the wall heat flux decreases with downstream position.The section components sit in a Delrin (acetal) frame, chosen for its low thermal conductivity and machinability. The wall-plate will be used to investigate thermal transport in non-equilibrium boundary layer flows. In this talk, we report on the validation tests performed to-date to investigate the aerodynamic and thermal performance of the wall-plate, and the capability of the controller to maintain the wall-plate at a pre-selected fixed temperature in steady and unsteady laminar boundary layer flow. [Preview Abstract] |
Monday, November 24, 2014 4:01PM - 4:14PM |
L26.00003: Log-Law scaling of a convective boundary layer in an unstably stratified turbulent channel flow Andrea Scagliarini, Halldor Einarsson, Armann Gylfason, Federico Toschi Turbulent convection is ubiquitous in a variety of natural and industrial flows. In particular, convective motions may play a role in sheared flows. In this work, we are concerned with the interplay of buoyancy and shear in the dynamical boundary layer structure. The lattice Boltzmann Method (LBM) is applied to study numerically an unstably-stratified, fully developed, turbulent channel flow, driven by a longitudinal pressure gradient and with an imposed transverse wall temperature difference along the direction of gravity. Spanning the friction Reynolds ($Re_{tau} \leq 205$) and Rayleigh numbers ($Ra \leq 1.3\times 10^7$) we could systematically study the influence of the convection on the boundary layer structure and mean profiles of flow quantities in the channel. Our focus is on providing physical understanding of the deviations observed from the logarithmic law of the wall due to the buoyant motions as well as providing a model of this behavior, and link with fundamental quantities of heat transfer in the convective channel flow. Our findings show that the introduction of an unstably stratified thermal field results in an effective drag increase in the channel flow, quantified in the logarithmic region by a modified log-law, with model parameters dependent on $Ra,Re_{tau}$. [Preview Abstract] |
Monday, November 24, 2014 4:14PM - 4:27PM |
L26.00004: Characteristics of Spatiotemporally Homogenized Boundary Layers at Atmospheric Reentry-like Conditions Rhys Ulerich, Robert Moser Turbulent boundary layers approximating those found on the NASA Orion Multi-Purpose Crew Vehicle thermal protection system during atmospheric reentry from the International Space Station have been studied by direct numerical simulation using a ``slow growth'' spatiotemporal homogenization approach recently developed by Topalian et al. The two data sets generated were $\mbox{Ma}_{e}\approx{}0.9$ and $1.15$ homogenized boundary layers possessing $\mbox{Re}_{\theta}\approx{}382$ and $531$, respectively. Edge-to-wall temperature ratios were approximately 4.15 and wall blowing velocities, $v_w^{+} = v_w / u_\tau$, were roughly $8\times10^{-3}$. The favorable pressure gradients had Pohlhausen parameters between 25 and 42. Nusselt numbers under 22 were observed. Small or negative displacement effects are evident. Near-wall vorticity fluctuations show qualitatively different profiles than observed by Spalart [J. Fluid Mech. 187 (1988)] or Guarini et al. [J. Fluid Mech. 414 (2000)] suggesting that the simulations have atypical structures perhaps as a consequence of wall blowing or the homogenization. [Preview Abstract] |
Monday, November 24, 2014 4:27PM - 4:40PM |
L26.00005: Effect of cooling on compressible wall-turbulence Andrew Trettel, Johan Larsson The modifications of the inner layer turbulence due to significant wall-cooling in a perfect gas is studied through a sequence of direct numerical simulations of compressible channel flow. The thickening of the buffer layer due to the imposed viscosity-gradient is quantified, as is the shift in the log-law intercept of the van Driest transformed velocity. The modification of the near-wall turbulent length scales and how these change with wall distance is examined. Finally, alternatives to the classic van Driest (1951) transformation of the mean velocity profile are considered. [Preview Abstract] |
Monday, November 24, 2014 4:40PM - 4:53PM |
L26.00006: Multilayer scaling of mean velocity and thermal fields of compressible turbulent boundary layer Weitao Bi, Bin Wu, Yousheng Zhang, Fazle Hussain, Zhen-Su She Recently, a symmetry based structural ensemble dynamics (SED) theory was proposed by She et al. for canonical wall bounded turbulent flows, yielding prediction of the mean velocity profile at an unprecedented accuracy (99\%). Here, we extend the theory to compressible turbulent boundary layers (TBL) at supersonic and hypersonic Mach numbers. The flows are acquired by spatially evolving direct numerical simulations (DNS). A momentum mixing length displays a four layer structure and quantitatively obeys the dilation group invariance as for the incompressible TBL. In addition, a temperature mixing length behaves very similarly to the momentum mixing length when the wall is adiabatic, with a small difference in the scaling exponents in the buffer layer - consistent with the strong Reynolds analogy. The Lie group based formulization of the two mixing lengths yields a multilayer model for the turbulent Prandtl number, along with predictions to the mean thermal and velocity profiles, both in good agreement with the DNS. Thus, we assert that the compressible TBLs are governed by the same symmetry principle as that in the canonical wall bounded turbulent flows, and its mean fields can be accurately described by the SED theory. [Preview Abstract] |
Monday, November 24, 2014 4:53PM - 5:06PM |
L26.00007: Turbulent heat flux measurements in thermally stable boundary layers Owen J. Williams, Tyler Van Buren, Alexander J. Smits Thermally stable turbulent boundary layers are prevalent in the polar regions and nocturnal atmospheric surface layer but heat and momentum flux measurements in such flow are often difficult. Here, a new method is employed using a nanoscale cold-wire (T-NSTAP) adjacent to a 2D PIV light sheet to measure these fluxes within rough-wall turbulent boundary layer. This method combines the advantages of fast thermal frequency response with measurement of the spatial variation of the velocity field. Resolution is limited solely by the separation of the probe and the light sheet. The new technique is used to examine the applicability of Monin-Obukhov similarity over a range of Richardson numbers from weak to strongly stable. In addition, the velocity fields are conditionally averaged subject to strong deviations of temperature above and below the local average in an effort to determine the relationship between the coherent turbulent motions and the fluctuating temperature field. [Preview Abstract] |
Monday, November 24, 2014 5:06PM - 5:19PM |
L26.00008: Thermal Transport Phenomena in the Quasi-laminarization Process of Turbulent Boundary Layers Luciano Castillo, Guillermo Araya, Fazle Hussain Direct Numerical Simulation of a spatially evolving turbulent boundary layer subject to strong favorable pressure gradient (SFPG) with eventual quasi-laminarization has been performed to include the thermal field. In this talk, the following questions will be addressed: i) to which extend the Reynolds analogy is satisfied during flow laminarization? ii) can the thermal boundary layer under quasi-laminarization be described as two quasi-independent inner/outer regions? To the best of our knowledge, documented investigation regarding heat transfer phenomena associated with quasi-laminarization process of spatially developing boundary layers is rather scarce. In order to introduce realistic thermal inlet fluctuations in a SFPG we employed the multi-scale technique for thermal boundary layers devised by Araya and Castillo PoF (2013). Such methodology enable us to more effective capture the pressure gradient and thermal field than using a single scaling approach. It is shown that the Reynolds normal stresses for the streamwise component remains frozen in space while the wall-normal and spanwise components continue to decrease as the flow moves downstream and never becomes laminar due to the survival of the upstream turbulence during dissipation on viscous time scales. [Preview Abstract] |
Monday, November 24, 2014 5:19PM - 5:32PM |
L26.00009: Effect of Small Roughness Elements on Thermal Statistics of Turbulent Boundary Layer at Moderate Reynolds Number Ali Doosttalab, Guillermo Araya, Ronald Adrian, Luciano Castillo DNS simulations of the zero pressure gradient turbulent boundary layer subject to forced convection over a transitionally rough surface with $k^+ \approx 11$ and Reynolds numbers based on momentum thickness of 2400, are presented for the first time. Prescribing realistic turbulent inlet conditions for rough surface and spatial evolving flows is extremely challenging. In order to solve the computationally intensive simulations, a dynamic method proposed by G. Araya \textit{et al. }(\textit{JFM, }vol. 670, pp. 581-605, 2011) for prescribing realistic inflow boundary conditions is used for simulations of spatially developing thermal turbulent boundary layers. Preliminary results showed how a transitionally rough surface alters thermal statistics in the inner and outer layers. Based on variations of $C_{f}$ and $St$, the validity of Reynolds analogy was tested and confirmed in the rough case and an increase to isotropy was obsereved. Furthermore, it was found that roughness enhances wall-normal heat flux transport in the inner layer and reduce it in the outer region of boundary layer. In addition. The rough surface decreased the ratio of Reynolds stress to turbulent heat flux in the near wall region, leading to a decreased turbulent Prandtl number. [Preview Abstract] |
Monday, November 24, 2014 5:32PM - 5:45PM |
L26.00010: An Integral Method to Evaluate Wall Heat Flux in Oscillatory Wall-Bounded Flow Alireza Ebadi, Christopher White, Ian Pond, Yves Dubief An integral method to evaluate wall heat flux in oscillatory wall-bounded flow is presented. The method is mathematically exact and has the advantage of having no explicit streamwise gradient terms. Importantly, the mathematical exactness of the method allows for a direct measurement of the wall heat flux without any a priori assumptions regarding the flow field or invoking flow transport analogies. It is useful in cases when measurements at multiple streamwise locations are not available or feasible, for flows with ill-defined outer boundary conditions, or when the measurement grid does not extend over the whole boundary layer thickness. The method is validated using DNS datasets of reciprocating turbulent channel flow with heat transfer for which independent estimates of wall heat flux were known, and the different results compare favorably. Complications owing to experimental limitations and measurement error in determining wall heat flux from the proposed method are presented, and mitigating strategies are described. [Preview Abstract] |
Monday, November 24, 2014 5:45PM - 5:58PM |
L26.00011: Connecting the classical limits: the Graetz-Nusselt problem for partial, homogeneous slip Rob Lammertink, Sander Haase, Jon Chapman, Peichun Tsai, Detlef Lohse The classical Graetz-Nusselt problem concerns the transport of heat between a hydrodynamically fully developed flow and the wall of a cylindrical pipe at constant temperature. In the thermally developing regime, the Nusselt number scales as Nu $\propto$ Gz$^{-\beta}$, where Gz $=$ RePr$D/L$ is the Graetz number. In case of a non-slippery wall $\beta = 1/3$, whereas for no-shear surfaces $\beta=1/2$. The generally assumed no-slip boundary condition does not always hold. Intrinsic slip lengths in micro- and nanofluidic systems vary from nearly zero to almost infinity. Here we studied the Graetz-Nusselt problem for partial slip. We present a solution for the Graetz-Nusselt problem for partial slip, connecting the two classical solutions. We show numerically and analytically that for surfaces displaying partial slip, $\beta$ gradually changes from $1/3$ to $1/2$. Also the developed Nusselt number Nu$_{\infty}$ slowly changes value from 3.66 to 5.78. We provide a mathematical and physical explanation for these two transitions points, which are separated more than one decade apart for $\beta$ and Nu$_{\infty}$. [Preview Abstract] |
Monday, November 24, 2014 5:58PM - 6:11PM |
L26.00012: Thermo-fluid-dynamics of turbulent boundary layer over a moving continuous flat sheet in a parallel free stream Bushra Afzal The momentum and thermal turbulent boundary layers over a continuous moving sheet subjected to a free stream have been analyzed in two layers (inner wall and outer wake) theory at large Reynolds number. The present work is based on open Reynolds equations of momentum and heat transfer without any closure model say, like eddy viscosity or mixing length etc. The matching of inner and outer layers has been carried out by Izakson-Millikan-Kolmogorov hypothesis. The matching for velocity and temperature profiles yields the logarithmic laws and power laws in overlap region of inner and outer layers, along with friction factor and heat transfer laws. The uniformly valid solution for velocity, Reynolds shear stress, temperature and thermal Reynolds heat flux have been proposed by introducing the outer wake functions due to momentum and thermal boundary layers. The comparison with experimental data for velocity profile, temperature profile, skin friction and heat transfer are presented. In outer non-linear layers, the lowest order momentum and thermal boundary layer equations have also been analyses by using eddy viscosity closure model, and results are compared with experimental data. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700