Bulletin of the American Physical Society
2006 59th Annual Meeting of the APS Division of Fluid Dynamics
Sunday–Tuesday, November 19–21, 2006; Tampa Bay, Florida
Session LL: Turbulent Boundary Layers V |
Hide Abstracts |
Chair: Alexander Smits, Princeton University Room: Tampa Marriott Waterside Hotel and Marina Meeting Room 8 |
Tuesday, November 21, 2006 8:00AM - 8:13AM |
LL.00001: The rough-wall turbulent boundary layer from the hydraulically smooth to the fully rough regime Michael Schultz, Karen Flack Turbulence measurements for rough-wall boundary layers are presented and compared to those for a smooth wall. The rough-wall experiments were made on a three-dimensional, rough surface geometrically similar to the honed pipe roughness used in the study of Shockling, Allen, {\&} Smits (2006). The present work is unique in that it covers a wide Reynolds number range (\textit{Re}$_{\theta }$ = 2,180 -- 27,100), spanning the hydraulically smooth to the fully rough flow regimes for a single surface, while maintaining a roughness height that is a very small fraction of the boundary layer thickness. In this investigation, the root-mean-square roughness height was at least three orders of magnitude smaller than the boundary layer thickness, and the K\'{a}rm\'{a}n number (\textit{$\delta $}$^{+})$, typifying the ratio of the largest to smallest turbulent scales in the flow, was as high as 10,100. The mean velocity profiles for the rough and smooth walls show remarkable similarity in the outer layer using velocity-defect scaling. The Reynolds stresses and higher order turbulence statistics also show excellent agreement in the outer layer. The results lend strong support to the concept of outer layer similarity for rough walls in which there is a large separation between the roughness length scale and the largest turbulence scales in the flow. [Preview Abstract] |
Tuesday, November 21, 2006 8:13AM - 8:26AM |
LL.00002: Turbulence measurements in a commercial steel pipe in the smooth to fully rough regime Richard Pepe, Alexander Smits Fully developed turbulent flow in a commercially rough pipe is studied using a crossed hot-wire probe. Streamwise and wall- normal turbulence components are obtained over a Reynolds number range from $1.1 \times 10^5$ to $9.8 \times 10^6$, covering the smooth to fully rough regime. Inner and outer scaling are applied to the turbulence intensity and spectra. The results are evaluated in terms of Perry's attached eddy model prediction and Townsend's `inactive' and `active' motions. [Preview Abstract] |
Tuesday, November 21, 2006 8:26AM - 8:39AM |
LL.00003: Mean velocity measurements in a commercial steel pipe in the smooth to fully rough regime Leif Langelandsvik, Gary Kunkel, Alexander Smits Mean velocity profiles and friction factors were obtained in fully-developed flow in a commercial, extruded steel pipe. The surface roughness was as delivered by the supplier, with $k_{rms} = 5 \mu$m. These are believed to be the first results ever obtained in commercial steel pipe under laboratory conditions. It was found that commercial steel pipe has a friction factor behavior that lies between the Colebrook type roughness, and sand-grain type roughness, in that the departure from the smooth pipe is rather sudden, with little or no evidence for the “inflectional” behavior characteristic of sand- grain roughness (and honed surface roughness). In addition, the equivalent sand-grain roughness of commercial steel pipe is closer to 2.0$k_{rms}$, rather than the commonly accepted value of 3.0$k_{rms}$. The velocity profiles in the rough regime show the expected departure below the logarithmic law, and they also follow Townsend's outer-layer similarity. [Preview Abstract] |
Tuesday, November 21, 2006 8:39AM - 8:52AM |
LL.00004: Rough-Wall Turbulent Boundary Layers James Allen, Jason Monty, Henry Ng, Min Chong In his 1962 report on ``Turbulent Boundary Layers in Incompressible Flow'' (Prog. Aero. Sci., Vol. 2, 1-219), J. C. Rotta identifies six possible turbulent boundary layers reaching precise equilirium. All bar one of these flows requires a roughened wall. At the University of Melbourne, a favourable pressure-gradient wind-tunnel has been carefully modified in an attempt to experimentally validate Rotta's fifth equilibrium condition. This particular flow requires exponentially increasing freestream velocity (in the streamwise direction) with constant height roughness elements on one wall. Although the difficulty of realising such a facility has been overcome, two inevitable hinderances of rough-wall boundary layer measurements remain: the uncertainty in virtual origin of the turbulent layer and the determination of wall shear stress. Fortunately, equilibrium flows afford an opportunity to indirectly determine wall shear stress through a momentum balance, provided the equilibrium condition of streamwise similarity of the mean velocity is fulfilled. The validity of determining wall shear stress as such has been investigated and compared with other commonly adopted methods. [Preview Abstract] |
Tuesday, November 21, 2006 8:52AM - 9:05AM |
LL.00005: Numerical Study of Rough and Smooth Turbulent Boundary Layers at Zero Pressure Gradient. Jorge Bailon-Cuba, Luciano Castillo The present study proposes an accurate numerical technique for determining the flow parameters of a rough turbulent boundary layer, based on the theory by George \& Castillo (GC-97). Moreover, an improvement in the Large Eddy Simulation (LES) of Bohr (2005) over a smooth flat plate, has been performed through a grid refinement and increase in the Reynolds number ($\delta^{+}$) range. This LES emphasizes a Rescaling-Recycling technique based on the Equilibrium Similarity Theory of GC-97, when it is implemented in the method originally proposed by Lund et al. (1998). The results, after comparing with the LES for smooth surfaces, and testing with experimental rough \& smooth data available, show that the ranges of the turbulent Reynolds number, $\delta^{+}$, and the blockage ratio, $k/\delta \ge 0.030$, at which similarity laws are expected to be valid are in consistency with the predictions by Jimenez (2005). The theoretical behavior of flow parameters such as $u_{\tau}$, $Re_{x}$, $Re_{\delta^{*}}$, $vs$ $Re_{\theta}$ indicate that for the rough surfaces tested, the GC-97 theory can be validated for hydraulically smooth, and transitionally rough surfaces until: $k^{+} \leq 35$. In addition, the analytical profiles of velocity ($U/U_{\infty}$), Reynolds shear stresses ($-\langle uv \rangle$), and Eddy viscosity ($\langle \nu_{T} \rangle$), are compared with the LES and experiments, showing good agreement (especially at high $\delta^{+}$'s values) in the inner and outer regions. [Preview Abstract] |
Tuesday, November 21, 2006 9:05AM - 9:18AM |
LL.00006: Large eddy simulation of turbulent pipe flows over surface roughness and their effects on the outer large structures Makoto Tsubokura, Jun Etoh, Shigeaki Kuroda Large Eddy Simulation (LES) of fully developed turbulent pipe flows with or without surface roughness were conducted using a very long streamwise analysis region to investigate the effects of surface roughness on the large-scale structures, which are typically observed in the logarithmic layer of wall turbulence. In a previous study we conducted an intensive grid resolution study to properly reproduce outer large-scale structures in plane channels and found that sufficiently fine grid resolutions of around $h_x^+\sim 30$ and $h_z^+\sim 20$ for streamwise and spanwise directions are required in the near wall region. Following this grid resolutions, our new pipe LES were carried out based on the fully conservative finite difference scheme in cylindrical coordinates along with the novel pole treatment developed recently by Morinishi et al. (2003). The roughness was made of ring-shaped slats with a square cross section, and modeled by the immersed boundary method. The effects are discussed in the context of the difference of the smooth and rough premultiplied spectra (both for streamwise and azimuthal) and related statistics at several roughness heights and longitudinal spacing as geometrical parameters. [Preview Abstract] |
Tuesday, November 21, 2006 9:18AM - 9:31AM |
LL.00007: Simulation of Turbulent Boundary Layer Flow with Large Roughness Erika Johnson, Chelakara Subramanian Several studies indicate that in situations where surface roughness is very strong, the friction velocity scaling for the mean and turbulent velocities are not satisfactory. Subramanian et al showed a dramatic effect of a strong irregular roughened surface on the turbulent properties. The log-law relation in the overlap region was distorted. A significant pressure gradient normal to the surface was observed with a concomitant increase in normal turbulent stress, $\overline {\mbox{v}^{'2}} $. The pressure gradient velocity scale, u$_{P}$, was suggested as a better alternative for capturing the effects of this roughness induced pressure gradient. Here, we performed a numerical simulation of a roughened boundary layer to gain more insight on the correlation between the wall normal pressure gradient and normal turbulent stress, $\overline {\mbox{v}^{'2}} $ and further validate this new pressure gradient velocity scale, u$_{P }$for different types of roughness. A two-dimensional flat plate computational model with strong regular (k-type) roughness, was constructed in GAMBIT and a CFD analysis performed using FLUENT, version 6.2. The roughness elements cause the pressure near the wall to increase suddenly at the first element and then decrease gradually similar to experiments. The variation of this normal pressure gradient is well correlated with the normal turbulent stress $\overline {\mbox{v}^{'2}} $ variation. Application of the u$_{P}$ as scaling parameter for other non-equilibrium flows such as with suction and blowing is also investigated. [Preview Abstract] |
Tuesday, November 21, 2006 9:31AM - 9:44AM |
LL.00008: The Character of Wall Turbulence in the Presence of Highly-Irregular Surface Roughness Yanhua Wu, Kenneth Christensen PIV measurements are made in the streamwise--wall-normal plane of a zero-pressure-gradient turbulent boundary layer over both smooth and rough ($\delta/k=25-28$; $\delta/k_s>40$) walls at $\mathrm{Re}_\theta=3900$ (transitionally rough) and $\mathrm {Re}_\theta=11000$ (fully rough). The roughness studied herein is replicated from surface scans of a turbine blade damaged by deposition of foreign materials and contains a broad range of topological scales. The mean velocity defect profiles as well as the Reynolds normal and shear stress profiles for the rough- wall flows collapse with the smooth-wall profiles in the outer region when scaled by their respective $u_\tau$ values. This collapse is consistent with Townsend’s wall similarity hypothesis. Quadrant decomposition of contributions to the mean Reynolds shear stress also reveal similarity between the smooth- and rough-wall flows. However, the two-point spatial velocity correlation coefficients appear to be more sensitive to the surface topology as the smooth- and rough-wall data show measurable differences. [Preview Abstract] |
Tuesday, November 21, 2006 9:44AM - 9:57AM |
LL.00009: Turbulence Structure in Rough and Smooth Wall Boundary Layers Ralph Volino, Michael Schultz, Karen Flack The outer region structure of turbulent boundary layers on smooth and rough walls was studied experimentally. Turbulence spectra were computed from LDV data. Velocity fields were computed from PIV data. Instantaneous swirl strength fields were computed from the velocity fields. The heads of hairpin vortices grouped as packets were visible in the streamwise wall normal plane, and the legs of these vortices were visible along the length of low speed streaks in streamwise spanwise planes at y/$\delta$=0.1 and 0.4. These structures, observed previously in smooth wall boundary layers, were qualitatively similar in the rough and smooth wall cases. Two point correlations of the velocity and swirl strength were quantitatively similar for the smooth and rough walls. The turbulence spectra and probability density functions of the turbulence and swirl strength also showed quantitative similarity between the rough and smooth wall cases when the results were normalized using the friction velocity and the boundary layer thickness. This similarity in turbulence structure is in agreement with the similarity in turbulence statistics reported previously. [Preview Abstract] |
Tuesday, November 21, 2006 9:57AM - 10:10AM |
LL.00010: Roughness Effects on Organized Motions in a Wall Shear Layer Flow Christian Haigermoser, Lukas Vesely, Massimillano Lapolla, Michele Onorato Turbulent boundary layer measurements on a zero-pressure gradient flat plate with two different roughness, a 2D and a 3D roughness, were carried out. The main object of the study was to investigate the impact of the wall roughness on the turbulent flow structures. The momentum thickness Reynolds number for the smooth wall was Re$_{\theta} \sim$ 1900. PIV measurements were taken in the streamwise wall-normal plane. The PIV images covered the whole logarithmic region and the major part of the outer layer. The instant flow images for the two roughness show features similar to the one expected in a smooth wall turbulent boundary layer, as described by Adrian et al. (JFM 2000). Statistical analysis was performed to enlighten quantitative differences between the different flow fields. For instance, two point streamwise velocity correlations show that the major effect of the roughness is to tilt the inclination of the hairpin vortex packets towards the wall normal direction; being the 3D roughness more effective in producing this displacement. Full results will be shown and discussed during the presentation. [Preview Abstract] |
Tuesday, November 21, 2006 10:10AM - 10:23AM |
LL.00011: The Effect of Surface Roughness on a Zero Pressure Gradient Turbulent Boundary Layer. Brian Brzek, Raul Bayoan Cal, Gunnar Johansson, Luciano Castillo A new set of experiments have been performed in order to study the effects of surface roughness on a zero pressure gradient turbulent boundary layer. The effect of roughness on the velocity deficit profiles and Reynolds stresses in outer variables is presented. It will be shown that for fixed experimental conditions (i.e., fixed upstream wind tunnel speed, trip wire, etc), the velocity deficit profiles collapse leading to self similar solutions. In addition, the Reynolds stresses show self similarity for fixed experimental conditions and are increasingly affected by the roughness as the roughness parameter, k$^{+}$, increased. Moreover, it is found that the shape of the Reynolds stress profiles are very different throughout the entire boundary layer, particularly for the $<$u$^{2}>$ component due to roughness. This is likely the result of the flow becoming more isotropic for increased k$^{+}$ and it will be seen in the analysis of the anisotropy tensor and the anisotropy invariant map. Moreover, increased production of $<$u$^{2}>$ and $<$uv$>$ due to roughness is also seen throughout the entire boundary layer although its overall role in the changing shape of the $<$u$^{2}>$ profiles still needs to be determined. [Preview Abstract] |
Tuesday, November 21, 2006 10:23AM - 10:36AM |
LL.00012: Effects of Surface Roughness on Favorable Pressure Gradient Turbulent Boundary Layers Raul Bayoan Cal, Brian Brzek, Gunnar Johansson, Luciano Castillo Different sets of experiments are carried out in order to
investigate the influences of the surface roughness on the
favorable pressure gradient (FPG) turbulent boundary layer.
Currently, there are no experiments reported on FPG turbulent
boundary layer over rough surfaces. This includes the Reynolds
stress measurements. The point at hand is to not only understand
the interaction between the rough surface and the outer flow, but
to also to include the external pressure gradient as the flow
evolves in the streamwise direction. These are obtained using
Laser Doppler Anemometry technique. It is found that the
roughness influences the velocity field in the outer flow. When
using the Zagarola and Smits scaling for the mean velocity
deficit profiles, a small difference due to the strength of the
pressure gradient is observed. In the Reynolds stresses, a
significant difference exists due to the rough surface and
Reynolds number effect in the rough FPG unlike the smooth FPG
data. Furthermore, the shape of the Reynolds stress in the
streamwise direction, $$, drastically changes in the
rough FPG case and is similar in the other components although
the magnitudes are different for the range of $k^+ =$ 35. The
$ |
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