Bulletin of the American Physical Society
73rd Annual Meeting of the APS Division of Fluid Dynamics
Volume 65, Number 13
Sunday–Tuesday, November 22–24, 2020; Virtual, CT (Chicago time)
Session Q05: Boundary Layers: Flow Over Roughness Elements (3:55pm - 4:40pm CST)Interactive On Demand
|
Hide Abstracts |
|
Q05.00001: Roughness height effects on turbulent boundary layers over rod-and cuboid-roughened walls Yun Kyung Choi, Hyeon Gyu Hwang, Young Mo Lee, Jae Hwa Lee Direct numerical simulations (DNSs) of spatially developing turbulent boundary layers (TBLs) over rod- and cuboid-roughened walls are conducted to investigate the effects of the roughness height on the flow characteristics in the outer layer. The rod and cuboid roughness height ($k)$ is varied in the range of 0.1$\le k$/$\theta_{in}\le $1.8 (13$\le \delta $/$k\le $285). As the roughness height increases, the roughness function ($\Delta U^{\mathrm{+}})$ increases and the magnitude of the Reynolds stresses in the outer layer also increases. The outer layer similarity between the flows over the rough and smooth-walls is established when $\delta $/$k \ge $ 250 and 100 for the 2D rod and 3D cuboid, respectively. The continuous increase of the Reynolds stresses in the outer layer with an increase of $k$/$\delta $is explained by a large population of very long structures over the rough-wall flows. Moreover, as $k$/$\delta $ increases, the wider characteristic width of the structure leads to frequent spanwise merging between adjacent structures. The active spanwise merging events between the spanwise-offset LSMs increase the appearance of meandering significantly as $k$/$\delta $ increases. [Preview Abstract] |
|
Q05.00002: Laminar skin friction drag reduction through the ``roller bearing effect'' over butterfly-inspired transverse cavities Sashank Gautam, Amy Lang, Leonardo Santos The Monarch butterfly wings are covered in minuscule scales (100 \textmu m) which align together in a pattern that resembles roof shingles. These scales are angled upwards such that transverse cavities form for a flow passing perpendicular to the rows of scales. As the air flow passes over the cavities, a single vortex is entrapped and forms at a very low Re (less than 10) inside each cavity. This embedded vortex acts as a fluidic bearing to the outer boundary layer and results in reduced skin friction drag. As the cavity geometry is varied to better mimic the butterfly scales, the shape of the vortex and the dividing streamline also changes, which in turn would affect the skin friction drag. This study aims to determine the cavity geometry that results in the optimum drag reduction. It is hypothesized that the slanted models inspired from observed butterfly scale geometries, with AR 2:1 with 45\textdegree cavity inclination angle and AR 3:1 with 22\textdegree inclination angle will result in higher drag reduction. This is because the slanted cavity vortex will be in less contact with cavity walls compared to rectangular cavity and thus be able to maintain a higher partial slip velocity as it interacts with the boundary layer. All experiments are conducted in a vertical tow-tank facility using DPIV. [Preview Abstract] |
|
Q05.00003: Restricted Nonlinear Comparisons of Turbulent Flows over Spanwise Heterogeneous Roughness Dennice Gayme, Benjamin Minnick, Xiaowei Zhu Turbulent flow over surfaces with spanwise heterogeneities is ubiquitous in both engineering applications and atmospheric science. These flows are represented in a variety of ways including spanwise-varying micro-grooves (riblets) or more abstractly as spanwise variations in roughness functions. Studies of these flows have revealed high and low momentum pathways influence secondary structures perpendicular to the mean flow that are related to the skin-friction drag generated. However, the nature of these pathways has been shown to vary across different studies. Here we do a detailed comparison of the two roughness models, using an immersed boundary method (IBM) to represent the riblets and an equilibrium log-law wall model in large eddy simulations (LES) with varying roughness heights. Our parametric study exploits the order reduction of the restricted nonlinear (RNL) methodology, which we first show reproduces the structures from respective direct numerical simulations (DNS) of flow over riblets and LES over spanwise variations in roughness. We then do a detailed comparison over a range of parameters to gain insight into the similarities and differences in flows arising through the two roughness models. [Preview Abstract] |
|
Q05.00004: Turbulent Secondary Flows in Boundary Layer Flows over Regular and Random Roughness Kristofer Womack, Ralph Volino, Michael Schultz, Charles Meneveau An experimental study was conducted on rough-wall, turbulent boundary layer flow with regular and random roughness element arrangements. Varying planform densities of truncated cone roughness elements in a square staggered pattern were investigated. The same planform densities were also investigated in random arrangements. Velocity statistics were measured via two-component laser Doppler velocimetry and stereo particle image velocimetry. A novel technique was used to determine friction velocity, $u_\tau$, from only the wall-normal profiles of mean streamwise velocity and Reynolds shear stress at a single streamwise location with accuracy comparable to force-balance measurements. Additionally, evidence is presented showing that the observed differences between regular and random surface flow parameters are due to the presence of low momentum pathways (LMPs) and high momentum pathways (HMPs) over the random surfaces which are not present over the regular surfaces. Previously identified spanwise heterogeneities thought to generate and sustain LMPs and HMPs in other studies do not seem to be present which suggests that these mechanisms are not necessary for LMP and HMP sustainment. Possible mechanisms for LMP and HMP self-sustainment are discussed. [Preview Abstract] |
|
Q05.00005: Restricted nonlinear (RNL) study of the mechanisms underlying drag reduction via riblets Xiaowei Zhu, Benjamin Minnick, Dennice Gayme Riblets in the form of spanwise-varying micro-grooves are known to change skin-friction drag in wall-bounded shear flows as a function of their size and shape. However, the drag reduction and degradation mechanisms associated with these flows are not fully understood. In this work, we seek to gain insight into this question through simulations of restricted nonlinear (RNL) flow over a range of riblet geometries. This reduced order model is first shown to reliably predict drag-reduction trends and the secondary flow motions attributed with the underlying mechanisms for different riblet shapes in low to moderate Reynolds number channel flow. The ability of the RNL model to reproduce quantities such as slip velocity, roughness function, and swirling strength, over this wide range of parameters suggests that its greatly reduced computational complexity can be exploited to perform parametric studies aimed at characterizing the underlying mechanisms. We use this model to isolate the contributions to the added stresses thought to contribute to increasing drag in certain parameter ranges. The reduced setting, which does not fully capture features such as the spanwise Kevin Helmholtz-like rollers, allows us to investigate how the absence of such mechanisms change the underlying behavior. [Preview Abstract] |
|
Q05.00006: A Simulation Study of the Flow Over a Roughness Element Ian J. Sysyn, Patrick H. Bonner, Frank G. Jacobitz A design focus of transportation systems is the reduction of aerodynamic drag forces in order to increase overall energy efficiency. An important component of such work is the transition of laminar to turbulent flows in a boundary layer. While laminar flows generally result in higher energy efficiencies, turbulent boundary layers can improve the stability of lift forces. This laminar to turbulent transition can be tripped by surface roughness, imperfections, or protrusions. The current study considers the flow around a cylindrical roughness element under laminar inflow conditions. The simulations aim to reproduce experiments (J. Lemarechal et al., 2018) visualizing the flow structure through the use of temperature-sensitive paint (TSP) applied to a heated surface. Both the experiments and simulations show the transition of the laminar boundary layer ahead of the roughness element to a more vortical flow state in its wake. Using Ansys CFD, the development of a horseshoe-shaped vortical structure as well as a recirculation zone directly downstream of the element is observed and the simulations qualitatively reproduce the experimental observations. The simulation results are also used to quantify the impact of the roughness element on the overall drag force and drag coefficient. [Preview Abstract] |
|
Q05.00007: Small wavy roughness effect on T-S wave and three-dimensional transition by Direct Numerical Simulation Shingo Hamada, Aiko Yakeno, Shigeru Obayashi, Bagus Nugroho Tani (1988) refered that the very small roughness with the height of lower than $6\nu/U_\tau$ could reduce skin friction drag, that scale was smaller than riblets (Walsh, 1982). In this study, we employ 2D and 3D DNS resolving each small roughness to analyze the effect of it on the turbulent transition of a flat plate boundary layer at Re$_{\delta_s}$ = 3535, where $\delta_s$ is the inlet boundary layer thickness. Artificial-disturbance of T-S wave instability is applied in an upstream. The roughness height is fixed at h$/\delta_s$ = 0.141, and the wavelength is varied from $\lambda/\delta_s$ = 0.89, 1.06, 1.35, 1.56, 2.91 and 5.80 (here we call it Case 1 to 6). Note that the Case 5 roughness frequency is set to correspond to T-S unstable frequency. Firstly our 2D simulation showed that the T-S wave growth was significantly restrained under a certain wavelength, a similar behavior was also observed by Tameike et al. (2020). Our 3D DNS showed that the T-S wave growth was almost identical to our 2D DNS. In Cases 3 and 4, the flow became three-dimensional more rapidly than in Cases 1, 2 and 5, and the phenomenon was considerably affected by the roughness shape. However, the entire transition-delay performance seemed to depend on the primary mode growth for the present cases. [Preview Abstract] |
|
Q05.00008: Three-dimensional flow structures and their interactions around a pair of cubic roughness elements embedded in the inner part of a turbulent channel flow Jian Gao, Joseph Katz The origin, evolution, and interactions of the 3D flow structures around a pair of roughness cubes with different spacings embedded in the inner part of a turbulent channel flow are measured using microscopic tomographic holography. The flow around each cube features open-type separation along the cube surfaces, and interactions among the vortical canopy covering the cube and multiple streamwise vortices. The incoming vorticity in the boundary layer and the cube front surface are major sources for the streamwise vorticity of the horseshoe and secondary vortices and for the wall-normal and spanwise vorticity of the canopy. Vortex stretching plays an important role in the interaction among these structures. Merging of the ``tip vortices'' developing along the cube upper edge, horseshoe and secondary vortices occur downstream at a distance that increases with the cube spacing. The spacing also affects the legs of the canopy, streamwise vortices along the inner side, dimensions of the separated regions, wall shear stresses, and the arch-type vortex behind the cube. The maximum wall shear stress occurs along the cube sides, between the horseshoe and the secondary streamwise vortices, owing to entrainment of high momentum flow towards the wall. [Preview Abstract] |
|
Q05.00009: Impact of Disturbances in the Overlying Atmospheric Boundary Layer on the Mechanisms of Urban Canyon Ventilation Tadeu Fagundes, Juan Ordonez, Neda Yaghoobian Increasing pollution levels and the resulting human health issues in urban areas are important problems that are linked to the capacity of ventilation and pollutant dispersion in urban streets. The transfer process and ventilation characteristics in urban streets are intrinsically related to the disturbances in the overlying atmospheric boundary layer. In this study, computational modelling was used to examine the dynamics between the flow within three-dimensional urban areas and the turbulent characteristics in the overlying flow. The features of the incoming boundary layer are controlled by altering the condition of the upstream landscape. The characteristics of turbulent flow in urban canyons are examined under different conditions, revealing distinct ventilation patterns and dispersion mechanisms that are deeply connected to the upstream conditions. In consequence, the air exchange capacity at the interface of the canyons is significantly altered. [Preview Abstract] |
|
Q05.00010: The Role of Roof Material in Diurnal Urban Air Quality: A Coupled Large-eddy Simulation and Surface Energy Balance Analysis Saurabh Saxena, Neda Yaghoobian Due to the increasing and adverse effects of the urban heat island phenomenon, heatwave events, and high urban air pollution, it has become increasingly important to have an in-depth understanding of the urban microclimates for mitigation purposes. The main objectives of this study are to analyze the relationship between building roof types/materials and the diurnal urban flow characteristics to assess the diurnal urban microclimates and street ventilation capacities. Through a series of novel computational simulations, involving detailed surface energy balance analyses coupled with large-eddy simulations, this study compares the thermo-fluid dynamics and pollution dispersion effects of asphalt, reflective, and green roofs. The results have shown significant differences in street ventilation capacity and pollution dispersion between the different roof-type cases. While the use of green roof reduces the urban air temperatures, a decline in the urban street air quality due to pollutant trapping has been observed at particular times throughout the day. [Preview Abstract] |
|
Q05.00011: Conjugate heat transfer in turbulent flows inside rough ducts Umberto Ciri, Kenan Wright, Stefano Leonardi Convective cooling inside ducts is a problem with significant practical importance. For instance, blades in gas turbine engines are machined with internal channels, where fluid spilled from the compressor cools the blades, externally impinged by hot fluid after combustion. Enhancement of the heat transfer coefficient for the flow inside the duct allows increasing the blade operative temperature, which improves the overall engine efficiency. Placing turbulators, such as V-shaped ribs, on the duct internal walls is a common technique to enhance heat transfer. The rough elements promote fluid mixing by increasing the turbulence intensity, leading to a heat exchange enhancement. This work aims to investigate the sensitivity of the thermal performance to the V-shaped ribs design parameters, such as the element pitch-to-height ratio. We will study the modifications induced by the ribs on turbulence, secondary motion and the impact on heat transfer coefficient and pressure drop. The study is conducted performing direct numerical simulations of conjugate heat transfer for turbulent flows inside ducts with rectangular V-shaped ribs on the wall. The duct walls and ribs are treated with the immersed boundary method, which allows us to efficiently resolve the conjugate heat transfer problem. [Preview Abstract] |
|
Q05.00012: Turbulent Transport of Firebrands in Urban Boundary Layers Iago Dal-Ri dos Santos, Neda Yaghoobian Generation of secondary fires by lofted burning materials (known as firebrands or flying embers) is an erratic process, in which wildland fires rapidly spread to unburned areas far downwind of the main fire-front. This phenomenon is the major mechanism of wildfire spread into fire-prone communities that poses serious threats to human lives and properties. The flight path and landing distribution of firebrands, and thus the spot fire risks depend heavily on the characteristics of ambient flow. By understanding the firebrand transport mechanisms in turbulent winds, it may be possible to better guide fire mitigation methods, reduce fire losses, and improve and guide fast fire-risk models. This study investigates the effect of topography-induced turbulence structures on firebrand transport, their smoldering lifetime, and spot fire risk over urban areas. Large Eddy Simulation is used to simulate the turbulent flow over arrangements of rectangular buildings. Evolving particles of different characteristics are released into the flow and their behavior is investigated in a Lagrangian framework. The fire hazard potentials and transport mechanisms are then analyzed based on the particle trajectories, their combustion state, and their landing distributions. [Preview Abstract] |
|
Q05.00013: The Effects of Surface Roughness on Turbulent Channel Flow Rong Ma, Karim Alame, Krishnan Mahesh Direct numerical simulation of flow in a turbulent channel with a random rough bottom wall is performed at friction Reynolds number 400 and 600. The rough surface corresponds to the experiments of Flack et al. (2019). The skin-friction coefficients are in agreement with the experimental results. DNS results are used to investigate roughness effects on dispersive flux, pressure fluctuations and formation of vortices in the roughness layer. The mean-square pressure fluctuations show an increase in the inner layer. The mechanism of the increased pressure fluctuations for rough-wall flows is investigated by examining the pressure Poisson equation source terms. The results show that the attached shear layer formed upstream of the protrusions is a primary source while the streamwise vortices in the troughs and in front of the protrusions also make a contribution. A computational capability is developed for global linear stability analysis of three-dimensional base flows. A time-stepper method is used in conjunction with the implicitly restarted Arnoldi iteration method. Validation and application to surface roughness will be discussed. [Preview Abstract] |
|
Q05.00014: Effect of wall roughness texture on transient turbulent channel flow Sai Chaitanya Mangavelli, Junlin Yuan, Giles Brereton Direct numerical simulation is used to study the effect of roughness topography on the response of channel-flow turbulence to an abrupt increase in bulk flow rate. Flows with two roughness topographies: a sand-grain roughness with a dominant wavelength; and a fractal roughness replicated from the surface of a hydraulic turbine blade, are compared with the baseline case of a smooth wall. The flow is accelerated from a bulk Reynolds Number of 3000 (transitionally rough) to 12000 (fully rough) over a very short time interval. The smooth-wall flow undergoes a reverse transition towards quasi-laminar flow. Wall roughness inhibits reverse transition and promotes a faster recovery to the new equilibrium state. The rough-wall responses are attributed to rapid growths in the pressure-strain rate and form-induced turbulence production in the early transient stage, with a dependency on surface topography. The turbine-blade roughness leads to a slower turbulence recovery compared to the sand-grain roughness. This difference is connected to the distribution of intense vortical motions, associated with the sparse high-slope roughness elements in turbine blade roughness. Other connections between the roughness topography and the turbulence response will be discussed. [Preview Abstract] |
|
Q05.00015: A simplified model for drag evaluation of a streamlined body with surface roughness Haoliang Yu, Umberto Ciri, Stefano Leonardi, Arif Malik Atmospheric particles, raindrops, hail, or sand can cause erosion of an airfoil, especially on the leading-edge (LE) region. As the erosion becomes more severe, the lifting profile can be gradually modified, leading to increased drag and decreased aerodynamic efficiency. In wind energy, for example, LE erosion of turbine blades can significantly reduce energy production. In the present study, surface imperfections on a streamlined body are idealized as forward-facing step(s) (FFS) for which the chordwise position, spanwise width, and distribution of the steps are varied. Direct numerical simulations (DNS) are performed to understand the interactions between three-dimensional FFS and the near-wall turbulent flows. The FFS and the streamlined body are defined by the ray triangle intersection test and modeled using immersed boundary method. A reduced order model (ROM) for efficient drag prediction is derived through a set of DNS with simplified geometrical variations. Performance of the ROM is assessed by applying it to a more realistic configuration that emulates randomly shaped LE erosion, and good agreement is obtained compared to the DNS results. [Preview Abstract] |
|
Q05.00016: Turbulent boundary layers over city-like roughness arrays Alexandros Makedonas, Marco Placidi, Mattreo Carpentieri By 2050, 68\% of the world’s population will live in cities. Understanding flow behaviour within and above cities is, therefore, of growing importance. To what extent do cities trap air pollution or increase the temperatures at street level, and do they affect the local weather cycles? Experimental models to study the effects of city geometry on flow behaviour in and above urban environments commonly use homogeneous-height roughness. However, cities are becoming increasingly vertically developed, and homogeneous height models are unlikely to portray an accurate picture of flow over real cities. Wind tunnel experiments were conducted at the University of Surrey over urban canopies to illustrate some essential differences of flow over uniform- and varied- height buildings. Flow characteristics of fully-developed boundary layers over four different rough surfaces that varied only in the standard deviation of element height were examined. The focus was on mean velocity profiles and higher-order quantities in the Inertial and the Roughness Sublayers. The presentation will cover the findings and draw comparisons between heterogeneous- and homogeneous- height roughnesses. [Preview Abstract] |
|
Q05.00017: Tall building wakes in isolation and in small clusters J. A. Minien, S. H. Choi, P. McDonald, S. Shone, P. Hayden, M. Placidi, Matteo Carpentieri, A. Robins The last few decades have witnessed a significant increase in the world's population accompanied by a process of urbanisation. To accommodate the population density, tall buildings, both in isolation and in clusters, have become ubiquitous. These severely affect the local environment, influencing air quality, wind loading, and urban energy demand. Wind tunnel experiments were carried out within the EnFlo laboratory at Surrey that focused on the wake flow downwind of tall buildings. Data both for a single tall building with varying aspect ratios and for small groups of tall buildings were investigated. In the latter dataset, we varied the number of buildings, their aspect ratio, and the spacing between them. The overall objective of this work is to investigate the wake's scaling laws to aid the modelling of this phenomenon. Measurements were taken predominately using two-component laser doper anemometry within a canonical atmospheric boundary layer created by a combination of Irwin spires and roughness elements. During the presentation, we will discuss the wake development/decay and compare these to simple models based on two- and three-dimensional wake theories. Finally, we will present appropriate scaling laws and lengthscales necessary to model the wake flow. [Preview Abstract] |
|
Q05.00018: Turbulent boundary layer over a grooved bump Edgardo Garcia, Eric Stout, Fazle Hussain The dynamics of a turbulent boundary layer over a bump due to passive surface modulations (i.e. longitudinal square grooves on a bump) are studied by DNS of a channel flow using the immersed boundary method. Mean (spanwise averaged) and phase-averaged flow statistics at several streamwise positions before, on, and after a bump are compared with those for a smooth bump. Surprisingly, the separated region on the rear of the bump is (z-averaged) 70{\%} larger due to reversed flow persisting within the grooves; for only above the grooves, the separated region is 15{\%} smaller. The peak turbulence production is shifted downstream 100 wall units and the magnitude of the peak decreases by 15{\%}. The grooves cause 0.5{\%} skin-friction drag reduction and 13{\%} form drag reduction as compared to the smooth bump. We find that upstream of the bump, flow enters the grooves and leaves them downstream; on the rear of the bump, the departure point in the grooves is upstream of the separation bubble on top of the grooves. This variation in the detachment point causes quasi-periodic streamwise vortical structures, which are shed from the corners of the grooves in the rear of the bump, that increase wall-normal momentum transport. These interactions between grooves and bumps enables better understanding of flows over sand dunes and mountains. [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