FREE ELECTRONIC LIBRARY - Dissertations, online materials

Pages:   || 2 | 3 |

«Characteristics of vortex packets in turbulent boundary layers By B H A R A T H R A M G A N A P A T H I S U B R A M A N I, E L L E N K. L O N G M I R ...»

-- [ Page 1 ] --


J. Fluid Mech. In Press.

Characteristics of vortex packets in

turbulent boundary layers

By B H A R A T H R A M G A N A P A T H I S U B R A M A N I,

E L L E N K. L O N G M I R E A N D I V A N M A R U S I C

Department of Aerospace Engineering and Mechanics, University of Minnesota,

107 Akerman Hall, 110 Union Street SE, Minneapolis, MN 55455, USA

e-mail : marusic@aem.umn.edu

(Received 07 August 2002 and in revised form 25 October 2002) Stereoscopic PIV was used to measure all three instantaneous components of the velocity field in streamwise-spanwise planes of a turbulent boundary layer at Reτ = 1060 (Reθ = 2500). Datasets were obtained in the logarithmic layer and beyond. The vector fields in the log layer (z + = 92 and 150) revealed signatures of vortex packets similar to those proposed by Adrian and co-workers in their PIV experiments. Groups of legs of hairpin vortices appeared to be coherently arranged along the streamwise direction. These regions also generated substantial Reynolds shear stress, sometimes as high as 40 times −uw. A feature extraction algorithm was developed to automate the identification and characterisation of these packets of hairpin vortices. Identified patches contributed 28% to −uw while occupying only 4% of the total area at z + = 92. At z + = 150, these patches occupied 4.5% of the total area while contributing 25% to −uw. Beyond the log layer (z + = 198 and 530), the spatial organisation into packets is seen to break down.

1. Introduction The structure of the turbulent boundary layer has been the subject of much research over the past 60 years because of its importance in practical applications. Some of the current studies are aimed at developing models to represent the boundary layer efficiently at higher Reynolds numbers. Various models have been proposed for a dominant structure including long quasi-streamwise vortices in the viscous buffer region (see Lyons, Hanratty & McLaughlin 1989 and Heist, Hanratty & Na 2000). However, a growing consensus is emerging for a model based on individual hairpin vortices (Theodorsen 1952) which is probably the simplest structural model that explains most of the features observed in wall turbulence. The existence of asymmetric hairpins in the region above the viscous layer was documented by Robinson (1991) in his analysis of early DNS results. Earlier, Offen & Kline (1975) used a hairpin model to explain the existence of ejections (Q2 events, u 0, w 0), volumes of low speed fluid that are pushed away from the wall and sweeps (Q4 events, u 0, w 0), volumes of faster fluid pushed towards the wall, in the logarithmic part of the boundary layer. In this paper u, v and w are defined as the fluctuating velocity components in the streamwise (x), spanwise (y) and wall-normal (z) directions respectively. Willmarth & Lu (1971) and Blackwelder & Kaplan (1976), among others, documented that ejections and sweeps make a substantial contribution to the Reynolds shear stress and therefore to the drag associated with the bounding surface.

Bandyopadhyay (1980) and Head & Bandyopadhyay (1981) proposed that hairpin vortices travel in groups after studying the flow visualisation experiments they performed 2 B. Ganapathisubramani, E. K. Longmire and I. Marusic on a zero pressure gradient boundary layer. Theoretical studies by Smith et al. (1991) and the study of DNS datasets at low Reynolds number (Reτ = δUτ /ν = 180, where δ is the boundary layer thickness, Uτ is the skin friction velocity and ν is the kinematic viscosity) performed by Zhou et al. (1999) demonstrated that a single vortex of sufficient circulation could spawn a trailing group of hairpins that convected at the same speed as the leading structure. More recently, Adrian, Meinhart & Tomkins (2000b), who performed particle image velocimetry (PIV) experiments in x−z planes of a zero pressure gradient boundary layer over a Reynolds number range of 355 Reτ 2000, identified packets of hairpin vortex heads that appeared regularly in the near-wall and logarithmic layers. Groups of 5hairpins, which extended over a length of 2δ, were determined to convect at a uniform streamwise velocity. The authors noted also that the presence of vortex packets explained the multiple ejections associated with individual bursts observed by Bogard & Tiederman (1986). The elongated zones of uniform streamwise velocity associated with the packets also help explain the long tails in two-point correlations of streamwise velocity found in Kovasznay, Kibens & Blackwelder (1970), Townsend (1976) and Brown & Thomas (1977).

Recently, Christensen & Adrian (2001) with PIV data in x − z planes of channel flow at Reτ = 547 and 1734, used linear stochastic estimation to estimate the conditionally averaged velocity field associated with swirling motion. They concluded that the mean structure consists of a series of swirling motions along a line inclined at 12◦ − 13◦ with the wall which is consistent with the earlier observations of packets of hairpin vortices.

An attached eddy model applied by Marusic (2001) also demonstrated that packets of eddies were required in order to match measured boundary layer turbulence statistics near the wall.

In recent work similar to the present study (Tomkins & Adrian 2002, see also the thesis by Tomkins 2001), PIV measurements of streamwise-spanwise planes of a turbulent boundary layer at 430 Reτ 2270 (only in-plane velocity components were measured) showed long low streamwise momentum zones enveloped by positive and negative vortex cores representing packets of hairpin structures.

The objective of the current study was to apply the stereo PIV technique to examine x−y planes in the outer region of a zero pressure gradient boundary layer (z + = zUτ /ν

60) in order to identify the typical structures and to determine their contribution to the Reynolds shear stress, thereby investigating the role, if any, the packet structures play in the transport of momentum in the boundary layers. Based on the previous work described above, feature identification algorithms were developed to search for individual hairpin vortices as well as packets of hairpins.

2. Facility and methods

2.1. Experimental facility, methods and qualification The measurements were carried out in a suction wind tunnel with working section of

0.33 m height, 1.22 m width and 4.8 m length. The coordinate system used is depicted in figure 1. Measurement planes were located 3.3 m downstream of a trip wire in a zeropressure-gradient flow with freestream velocity U∞ = 5.9 m s−1. Hot-wire measurements showed that the turbulence intensity in the freestream was less than 0.2%. The wall shear stress (τw ) was computed using the mean velocity profile and the Clauser chart method.

All quantities measured and computed are normalised using the skin friction velocity Uτ (= τw /ρ, where ρ is the density of the fluid) and ν and are denoted with a superscript +. The Reynolds number based on the momentum thickness Reθ was 2500, and Reτ was

1060. The value of δ in the region of the measurement planes was 69 mm.

Vortex packets in boundary layers 3

Figure 1. Experimental facility.

The flow was seeded with olive oil droplets (size ∼ 1 µm) that were generated by eight Laskin nozzle units set up in parallel. The oil droplets were ingested into the intake of the wind tunnel upstream of honeycomb straighteners and screens used for flow conditioning.

At the test section, glass side-walls and a glass bottom wall were installed in the wind tunnel to provide high-quality optical access. The seed particles were illuminated by pulsed sheets from two Nd:YAG lasers (Big Sky CFR200) directed through one side window and oriented parallel with the bottom wall of the tunnel. (See figure 1). The laser pulse energy was 120 mJ, and the thickness of each sheet was 0.3 mm. Sets of digital images were captured by two Kodak Megaplus CCD cameras (1024 × 1024 pixels) at z + = 92, 198 (z/δ = 0.2) and 530 (z/δ = 0.5) and by TSI Powerview 2048 × 2048 pixel resolution cameras at z + = 150. Nikon Micro Nikkor 60 mm f/2.8 lenses were used with both camera types. A TSI synchroniser box controlled the strobing and timing of the cameras and lasers. The dual frame acquisition rate was 15 Hz. The cameras were aligned in a plane parallel with the x − y flow plane and inclined at angles of 15◦ with the z axis as shown in figure 1. Additional details of the calibration, experimental arrangement and vector reconstruction schemes are given in Ganapathisubramani, Longmire & Marusic (2002).

For all derived quantities, the interrogation spot size used was 16×16 pixels (∼ 20×20 wall units) with 50% overlap. Hence the spacing between adjacent vectors in either direction is 10 wall units (∼ 0.65 mm). The measurements had two main sources of uncertainty.

First, the Gaussian peak fit in the cross-correlation algorithm generated an uncertainty of approximately 0.1 pixels or 0.016U (where U is the mean velocity in the streamwise direction). Second, a residual error arose due to the least-square curve-fit in solving the four pixel-displacement equations in three unknowns (see Ganapathisubramani et al.

2002 for details). This residual error can play a major role in the uncertainty if concentration gradients in seeding occur within the flow field. In this experiment, such gradients were not significant, and this error was on average about half of the Gaussian error. The vector fields from the images acquired with the Megaplus cameras covered an area of

1.2δ × 1.2δ. For the Powerview cameras, the field size was approximately 2.4δ × 2.4δ making the spacing between vectors similar to that in the other planes. The larger area was acquired by moving the plane of the cameras away from the object plane.

Velocity gradients were computed from vector fields using a second order central difB. Ganapathisubramani, E. K. Longmire and I. Marusic

–  –  –

ference scheme wherever possible in the domain and a first order forward or backward difference at the boundaries. To identify swirling motion caused by the eddies in the flow field, we used swirl strength λci (Zhou et al. 1999), which is defined as the magnitude of the imaginary part of the eigenvalue of the local velocity gradient tensor. Since the PIV images were planar, we used only the in-plane gradients and formed a two dimensional form of the tensor. Vortices could be identified by extracting iso-regions of λci (Adrian, Christensen & Liu 2000a).

The values of U computed from the PIV data at each wall-normal location agreed well with the hot-wire measurements. Table 1 shows a summary of some ensemble averaged turbulence statistics. The values shown compare well with previous data in the literature (Balint, Wallace & Vukoslavcevic 1991; Naguib & Wark 1992; Adrian et al. 2000b).

2.2. Feature extraction Although a variety of coherent structures can be found in a turbulent boundary layer, we focus on the hairpin vortices discussed in the introduction. Since, the stereo PIV data are obtained in streamwise-spanwise planes, hairpin legs could appear as neighbouring cores of positive and negative wall-normal vorticity (ωz ) aligned in the spanwise direction.

If these hairpins were to travel in groups, one might also see multiple pairs of cores aligned approximately along the streamwise direction. Also, individual or groups of angled hairpins are expected to generate large values of instantaneous Reynolds shear stress (−uw+ = −uw/Uτ 2 ). Hence, an algorithm was developed to identify regions in which all of these events occur. In the algorithm description, the flow is assumed to be from left

to right :

• Step 1: Regions of positive vorticity 1.5σωz + and negative vorticity −1.5σωz + are detected. At z + = 92, this value was ±0.065.

• Step 2: Zones of strong Reynolds shear stress (−uw+ ) greater than 2σ−uw + (where σ−uw + is the r.m.s of −uw+ ) are detected independently. No distinction is attached to sweep or ejection events. The values of σ−uw + at z + = 92 and 150 were 3.43 and 2.68 respectively.

• Step 3: All the zones found in step 2 are searched to mark regions of high Reynolds shear stress occurring between strips of positive and negative vorticity (found in step

1) such that positive vorticity lies beneath negative vorticity. This vorticity arrangement was chosen to fit the pattern displayed by the legs of a hairpin vortex.

• Step 4: Each location of large −uw+ marked in step 3 is used as a seed point for a region growing algorithm. This algorithm, based on the streamwise velocity, identifies a connected region of neighbouring points lying within local thresholds such that the final coherent region is a patch of uniform momentum. The lower bound of the velocity was fixed as the minimum velocity in the Reynolds shear stress zone and the upper bound was the average velocity in the two surrounding vorticity regions. If multiple seed zones are enveloped by the same vorticity regions, the thresholds are adjusted such that the Vortex packets in boundary layers 5 lower bound is the minimum velocity in all the seed points of −uw+ in that region. If the region growth from one location intersects the patches from other seed points, then they are merged and marked as one patch.

• Step 5: After detection of all patches in a given vector field, patches that are adjacent to one another along the streamwise direction are merged provided that the streamwise distance between the patches is less than the average width of the patches being merged.

3. Results and discussion Please note that the flow is from left to right in all vector plots, and the local mean U is subtracted from the vectors to clearly illustrate the slow and fast moving zones. The contour/vector plots shown in figure 2 are from individual realisations, but the patterns are representative of those found in many vector fields. The plots revealing the statistical features of the hairpin packets (figure 3) were obtained by investigating all vector fields at a given wall normal position.

Pages:   || 2 | 3 |

Similar works:

«Determination of Texture From Individual Grain Orientation Measurements John E. Blendell, Mark D. Vaudin and Edwin R. Fuller, Jr. Ceramics Division Materials Science and Engineering Laboratory National Institute of Standards and Technology Gaithersburg, MD 20899 USA Abstract We present a technique for determining the texture of a polycrystalline material based on the measurement of the orientation of a number of individual grains. We assumed that the sample has fiber (i.e. axisymmetric) texture...»

«CHAPTER 1 NANO-CMOS SCALING PROBLEMS AND IMPLICATIONS 1.1 DESIGN METHODOLOGY IN THE NANO-CMOS ERA As process technology scales beyond 100-nm feature sizes, for functional and high-yielding silicon the traditional design approach needs to be modified to cope with the increased process variation, interconnect processing difficulties, and other newly exacerbated physical effects. The scaling of gate oxide (Figure 1.1) in the nano-CMOS regime results in a significant increase in gate direct...»

«How to Reverse Your Cataracts Naturally. 5 Ways to Do It How to Reverse Your Cataracts Naturally: 5 Ways to Do It William Bodri The Skeptical Nutritionist Naturopathic Educator 1 How to Reverse Your Cataracts Naturally. 5 Ways to Do It Copyright Copyright © 2003, William Bodri All rights reserved in all media First edition 2003. Top Shape Publishing, LLC 1135 Terminal Way Suite 209 Reno, Nevada 89502 No part of this book may be reproduced, stored in a retrieval system, or transmitted by any...»

«Automatic Extraction of Generic House Roofs from High Resolution Aerial Imagery ? Frank Bignone, Olof Henricsson, Pascal Fua+ and Markus Stricker Communications Technology Laboratory Swiss Federal Institute of Technology ETH CH-8092 Zurich, Switzerland + SRI International, Menlo Park, CA 94025, USA Abstract. We present a technique to extract complex suburban roofs from sets of aerial images. Because we combine 2-D edge information, photometric and chromatic attributes and 3-D information, we...»



«ECNDT 2006 Tu.2.1.1 In-Service Inspection Concept for GLARE® – An Example for the Use of New UT Array Inspection Systems Wolfgang BISLE, Theodor MEIER, Sascha MUELLER, Sylvia RUECKERT Airbus, Bremen Abstract. UT phased arrays and array transducers for FIT (Field Inspection Technology) open new chances for NDT of complex structure materials in aeronautics. GLARE® is such a new material which will widely be used on the new Airbus A380. Even as GLARE® in the A380 needs no scheduled NDT...»

«Otty Lake Association Annual General Meeting (AGM) Minutes Saturday, July 9th, 2016, 9:30 – 11:30 am The Perth Legion Welcome – Reid welcomed everyone to the AGM at 9:30 am. He thanked the displayers. He also thanked Cathy Buffam and Tiger Tanner for all their work as the AGM’s technical team for many years. He thanked Gail Read and her daughter Lana, Krista Hearty McLean and Chuck Shenkman for taking over as our technical team this year. In addition to celebrating the 200th anniversary...»

«3 Whose Apple Is It Anyway! Copyright © 2014 Linda F. Williams. All rights reserved. No portion of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means — electronic, mechanical, photocopy, recording, scanning, or other — except for brief quotations in critical reviews or articles, or as specifically allowed by the U. S. Copyright Act of 1976, as amended, without the prior written permission of the publisher. Published by Whose Apple Press...»

«Martin Potter NATURE IN MODERNITY: CAN IT SIGNIFY? – DAVID JONES AND NATURAL OBJECTS AS SIGNS Keywords: nature, symbol, poetry, utility aesthetics, modernity, ‘the Break’. Abstract: Since the Enlightenment nature has often been seen as a resource for exploitation rather than as an object of wonder. A combination of utilitarian modes of thought, the growth in the prestige of science and technology, and processes of industrialization, has favoured the tendency to view natural objects as...»

«Installation Manual Frameworx Gutter and Capping August 2012 / 84-900030-000 Frameworx Gutter and Capping Installation Manual © August 2012 by the Brunswick Bowling and Billiards Corporation. All rights reserved. A-2, BallWall, Frameworx, GS-Series, Pinball Wizard, and Tel-E-foul, and are registered trademarks of the Brunswick Bowling and Billiards Corporation. Reorder Part No. 84-900030-000 Notice: If available, updates to this manual can be found on-line at www.centermaster.com. All...»

«CURRICULUM VITAE YITZHAK FRIED Work Address: Home Address: Management Department 10 D Kings Court Whitman School of Management Camillus, NY 13031 Syracuse University Phone: 315-487-1264 Phone: (315) 443-3639 E-mail: yfried@syr.edu Education Ph.D. Institute of Labor and Industrial Relations, University of Illinois at Urbana-Champaign, 1985 M.A. Tel Aviv University (Israel), 1979 B.A. Bar-Ilan University (Israel), 1976 Work Experience August 2004presentProfessor, Management Department, Whitman...»

<<  HOME   |    CONTACTS
2016 www.dissertation.xlibx.info - Dissertations, online materials

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.