Bulletin of the American Physical Society
69th Annual Meeting of the APS Division of Fluid Dynamics
Volume 61, Number 20
Sunday–Tuesday, November 20–22, 2016; Portland, Oregon
Session M33: Turbulence: Flow Thru Pipes |
Hide Abstracts |
Chair: Pinaki Chakraborty, Okinawa Institute of Science and Technology Room: Oregon Ballroom 202 |
Tuesday, November 22, 2016 8:00AM - 8:13AM |
M33.00001: Turbulence in Reynolds' flashes Rory Cerbus, Chien-chia Liu, Gustavo Gioia, Pinaki Chakraborty Osborne Reynolds' seminal work from 1883 revealed that the transition from quiescent, laminar flow to a turbulent pipe filled with roiling eddies is mediated by localized flashes of fluctuations. Later work has unveiled many features of these flashes: they proliferate or fade away, maintain their shape or continually expand. The nature of the fluctuations in the flashes, however, has remained mysterious. Here, using measures traditionally attributed to high Reynolds number (Re) flows, we present experimental results on the fluctuations of the flashes. Our results suggest that the transition to turbulence is the low Re limit of the high Re, fully developed flow. [Preview Abstract] |
Tuesday, November 22, 2016 8:13AM - 8:26AM |
M33.00002: Self-similarity of the large-scale motions in turbulent pipe flow Leo Hellstr\"om, Ivan Marusic, Alexander Smits Townsend's attached eddy hypothesis assumes the existence of a set of energetic and geometrically self-similar eddies in the logarithmic layer in wall-bounded turbulent flows. These eddies can be completely scaled with the distance from their center to the wall. We performed stereo PIV measurements together with a proper orthogonal decomposition (POD) analysis, to address the self-similarity of the energetic motions, or eddies, in fully-developed turbulent pipe flow. The resulting modes/eddies, extracted at $Re_\tau = 2460$, show a self-similar behavior for eddies with wall-normal length scales spanning a decade. This single length scale provides a complete description of the cross-sectional shape of the self-similar eddies. [Preview Abstract] |
Tuesday, November 22, 2016 8:26AM - 8:39AM |
M33.00003: High Reynolds number examination for fully developed pipe flow - Mean velocity profile and friction factor Noriyuki Furuichi, Yuki Wada, Yoshiyuki Tsuji, Yoshiya Terao The pipe flow examinations at high Reynolds numbers up to Re$=$18,000,000 are performed using the high Reynolds number actual flow facility “Hi-Reff” at AIST, NMIJ. The precise measurements of the friction factor and velocity profile are achieved by the highly accurate measurement of the flow rate. The friction factor data is obviously different from the Prandtl equation and the experimental results from the Superpipe at Princeton University. The deviation from the Superpipe is -6{\%} at Re$=$10,000,000. The velocity profile is measured by LDV. The consistency between the mean velocity profile and the friction factor measured is investigated. The velocity profile data is fitted to a velocity profile form based on the log law, and an equation for the friction factor is derived by integration. The derived equation for the friction factor accurately represents the friction factor data. The deviation from the friction factor data is less than 1{\%}. Based on the equations for the friction factor derived using the mean velocity profile, the best-fitting constants for the friction factor data are also proposed. For the high Reynolds number region, the Karman constant given by the velocity profile and the friction factor is completely consistent and it is 0.383. [Preview Abstract] |
Tuesday, November 22, 2016 8:39AM - 8:52AM |
M33.00004: Statistical growths of turbulent structures in a pipe flow. Junsun Ahn, Hyung Jin Sung The streamwise and spanwise (or azimuthal) growths of turbulent coherent structures in a turbulent pipe flow (Re$_{\mathrm{\tau }}=$3008) are explored. Two-point correlation and 1-D pre-multiplied energy spectra of the streamwise velocity fluctuations are obtained to analyze the statistical growths of the streamwise and spanwise structures. The streamwise and spanwise length scales linearly grow along the wall-normal distance and the relationship between both length scales is shown to be linear, which support the attached eddy hypothesis. Furthermore, the statistical scalings of the coherent structures are demonstrated and compared to 2-D pre-multiplied energy spectra of the streamwise velocity fluctuations. Finally, the relationship between the streamwise and spanwise structures is analyzed by using the POD based on the translational invariance method (Duggleby et al. 2009). Several representative energetic modes are observed. The combinations of the energetic modes are used to examine the behaviors of the large- and very-large-scale motions. [Preview Abstract] |
Tuesday, November 22, 2016 8:52AM - 9:05AM |
M33.00005: Five layers in a turbulent pipe flow. Jinyoung Lee, Junsun Ahn, Hyung Jin Sung The scaling laws governing the five layers of the mean velocity distribution of a turbulent pipe flow were characterized using the available DNS data (Re$_{\mathrm{\tau }} \quad =$ 544, 934, 3008). Excluding the very near-wall and core regions, the buffer, meso- and log layers were identified by examining the streamwise mean momentum equation and the net force spectra. The (outer) log layer was located in the overlap region where the viscous force was negligible. Another (inner) log layer was observed in the buffer layer, in which the viscous force was directly counterbalanced by the turbulent inertia. A meso-layer between the buffer and outer log layers was found to feature viscous effects. The acceleration force of the large-scale motions (LSMs) penetrated the outer log layer at higher Reynolds numbers, as observed in the net force spectra. The acceleration force of the LSMs became strong and was counterbalanced by the deceleration force of the small-scale motions (SSMs), indicating that the inner and outer length scales contributed equally to the meso-layer. The outer log layer was established by forming an extended connection link between the meso- and outer layers. [Preview Abstract] |
Tuesday, November 22, 2016 9:05AM - 9:18AM |
M33.00006: Effect of Reynolds number on flow and mass transfer characteristics of a 90 degree elbow Nobuyuki Fujisawa, Yuya Ikarashi, Takayuki Yamagata, Syoichi Taguchi The flow and mass transfer characteristics of a 90 degree elbow was studied experimentally by using the mass transfer measurement by plaster dissolution method, the surface flow visualization by oil film method and stereo PIV measurement. The experiments are carried out in a water tunnel of a circular pipe of 56mm in diameter with a working fluid of water. The Reynolds number was varied from 30000 to 200000. The experimental result indicated the change of the mass transfer coefficient distribution in the elbow with increasing the Reynolds number. This phenomenon is further examined by the surface flow visualization and measurement of secondary flow pattern in the elbow, and the results showed the suggested change of the secondary flow pattern in the elbow with increasing the Reynolds numbers. [Preview Abstract] |
Tuesday, November 22, 2016 9:18AM - 9:31AM |
M33.00007: Turbulent pipe flows subjected to temporal decelerations Wongwan Jeong, HyungJae Lim, Jae Hwa Lee Direct numerical simulations of temporally decelerating turbulent pipe flows were performed to examine effects of temporal decelerations on turbulence. The simulations were started with a fully developed turbulent pipe flow at a Reynolds number, $Re_{D}$=24380, based on the pipe radius ($R$) and the laminar centerline velocity ($U_{c0}$). Three different temporal decelerations were imposed to the initial flow with $f$=$|dU_{b}/dt|$=0.00127, 0.00625 and 0.025, where $U_{b}$ is the bulk mean velocity. Comparison of Reynolds stresses and turbulent production terms with those for steady flow at a similar Reynolds number showed that turbulence is highly intensified with increasing $f$ due to delay effects. Furthermore, inspection of the Reynolds shear stress profiles showed that strong second- and fourth-quadrant Reynolds shear stresses are greatly increased, while first- and third-quadrant components are also increased. Decomposition of streamwise Reynolds normal stress with streamwise cutoff wavelength ($\lambda_{x}$) 1$R$ revealed that the turbulence delay is dominantly originated from delay of strong large-scale turbulent structures in the outer layer, although small-scale motions throughout the wall layer adjusted more rapidly to the temporal decelerations. [Preview Abstract] |
Tuesday, November 22, 2016 9:31AM - 9:44AM |
M33.00008: Transition in Pulsatile Pipe Flow Pavlos Vlachos, Melissa Brindise Transition has been observed to occur in the aorta, and stenotic vessels, where pulsatile flow exists. However, few studies have investigated the characteristics and effects of transition in oscillating or pulsatile flow and none have utilized a physiological waveform. In this work, we explore transition in pipe flow using three pulsatile waveforms which all maintain the same mean and maximum flow rates and range to zero flow, as is physiologically typical. Velocity fields were obtained using planar particle image velocimetry for each pulsatile waveform at six mean Reynolds numbers ranging between 500 and 4000. Turbulent statistics including turbulent kinetic energy (TKE) and Reynolds stresses were computed. Quadrant analysis was used to identify characteristics of the production and dissipation of turbulence. Coherent structures were identified using the $\lambda _{\mathrm{ci}}$ method. We developed a wavelet-Hilbert time-frequency analysis method to identify high frequency structures and compared these to the coherent structures. The results of this study demonstrate that the different pulsatile waveforms induce different levels of TKE and high frequency structures, suggesting that the rates of acceleration and deceleration influence the onset and development of transition. [Preview Abstract] |
Tuesday, November 22, 2016 9:44AM - 9:57AM |
M33.00009: Numerical simulation of pulsatile flow in rough pipes Cheng Chin, Jason Monty, Andrew Ooi, Simon Illingworth, Ivan Marusic, Alex Skvortsov Direct numerical simulation (DNS) of pulsatile turbulent pipe flow is carried out over three-dimensional sinusoidal surfaces mimicking surface roughness. The simulations are performed at a mean Reynolds number of $Re_\tau \approx$ 540 (based on friction velocity, $u_\tau$, and pipe radii, $\delta$) and at various roughness profiles following the study of Chan et al. (JFM 771, 743 - 777, 2015), where the size of the roughness (roughness semi-amplitude height $h^+$ and wavelength $\lambda^+$) is increased geometrically while maintaining the height-to-wavelength ratio of the sinusoidal roughness element. Results from the pulsatile simulations are compared with non-pulsatile simulations to investigate the effects of pulsation on the Hama roughness function, $\Delta U^+$. Other turbulence statistics including mean turbulence intensities, Reynolds stresses and energy spectra are analysed. In addition, instantaneous phase (eg. at maximum and minimum flow velocities) and phase-averaged flow structures are presented and discussed. [Preview Abstract] |
Tuesday, November 22, 2016 9:57AM - 10:10AM |
M33.00010: Generation of Turbulent Inflow Conditions for Pipe Flow via an Annular Ribbed Turbulator Nima Moallemi, Joshua Brinkerhoff The generation of turbulent inflow conditions adds significant computational expense to direct numerical simulations (DNS) of turbulent pipe flows. Typical approaches involve introducing boxes of isotropic turbulence to the velocity field at the inlet of the pipe. In the present study, an alternative method is proposed that incurs a lower computational cost and allows the anisotropy observed in pipe turbulence to be physically captured. The method is based on a periodic DNS of a ribbed turbulator upstream of the inlet boundary of the pipe. The Reynolds number based on the bulk velocity and pipe diameter is 5300 and the blockage ratio (BR) is 0.06 based on the rib height and pipe diameter. The pitch ratio is defined as the ratio of rib streamwise spacing to rib height and is varied between 1.7 and 5.0. The generation of turbulent flow structures downstream of the ribbed turbulator are identified and discussed. Suitability of this method for accurate representation of turbulent inflow conditions is assessed through comparison of the turbulent mean properties, fluctuations, Reynolds stress profiles, and spectra with published pipe flow DNS studies. The DNS results achieve excellent agreement with the numerical and experimental data available in the literature. [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. |
© 2021 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
1 Research Road, Ridge, NY 11961-2701
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700