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 M27: Turbulence Theory: Measurements |
Hide Abstracts |
Chair: Lester Su, Stanford University Room: 2009 |
Tuesday, November 25, 2014 8:00AM - 8:13AM |
M27.00001: Full 3D derivative moments from an L-shaped SPIV experiment in the near wall region Jean-Marc Foucaut, Christophe Cuvier, Michel Stanislas Stereoscopic PIV is now a relevant method to measure turbulent flow. This method allows the measurement of the three components of the velocity in a plane with an accuracy of about 1-2{\%}. For turbulent flows usually only the large scale motions are investigated due to the limited spatial resolution of the PIV. The main difficulty comes when the derivative has to be computed due to the noise (Foucaut, 2002). The present communication proposes a method to determine the derivative moments which combines both the derivative and the statistic computation from a specific SPIV experiment. Balint et al (1991) published experimental results of derivative moments obtained by HWA in a turbulent boundary layer. A special multiwire probe was designed for this measurement. These results are globally of the same order as the DNS (Antonia et al. 1991), except for the derivative of the streamwise component for which Taylor's hypothesis was used. This was not necessary with stereoscopic PIV. The experiment was performed in the LML 20 m boundary layer facility to determine all the derivative moments needed to determine the dissipation. The Reynolds number was Re$_{\theta}$ $=$ 7500. Measurements were taken in two normal planes in order to compute all the derivatives of the three components. For the PIV there is a trade-off between field of view and the interrogation window sizes, so the derivative filter choice and the measurement noise management are particularly discussed. [Preview Abstract] |
Tuesday, November 25, 2014 8:13AM - 8:26AM |
M27.00002: Results of experimental investigation of the dissipation rate in the near wall region William George, Jean-Marc Foucaut, Christophe Cuvier, Michel Stanislas The present idea is to propose a method to determine the dissipation rate from a specific SPIV experiment: $\varepsilon = \nu\left[\left\langle \frac{\partial u_i}{\partial x_j} \frac{\partial u_i}{\partial x_j} \right\rangle + \left\langle \frac{\partial u_i}{\partial x_j} \frac{\partial u_j}{\partial x_i} \right\rangle \right]$ is strongly linked to the small scales of a turbulent flow. It is indispensable for turbulence modelling. Yet it has seldom been measured. To obtain the full dissipation it is necessary to measure the full instantaneous gradient tensor and to compute the 15 moments. But, all are difficult to measure accurately. George and Hussain (1991) proposed to simplify this computation by using different hypotheses in order to reduce the number of terms. Their hypotheses were local homogeneity and local axisymmetry, both of which are more general than the usual assumptions of local isotropy. An experiment was performed in the LML boundary layer facility to determine all the necessary derivative moments. A detailed analysis of the errors in derivative measurements was carried out (see Foucaut et al. 2014 APS), as well as applying and using consistency checks derived from the continuity equation and local homogeneity. Local homogeneity estimates of the dissipation are accurate everywhere to within a few percent. Both local axisymmetry and local isotropy work almost as well outside of y$+ \quad =$ 100, but only local axisymmetry provides a reasonable estimate closer to the wall. The results are remarkably similar to those of Antonia et al (1991) from DNS results of a channel flow at low Reynolds number. [Preview Abstract] |
Tuesday, November 25, 2014 8:26AM - 8:39AM |
M27.00003: Direct observation of energy cascade in three-dimensional turbulence Haitao Xu, Fabio Di Lorenzo, Alain Pumir, Eberhard Bodenschatz In three-dimensional turbulence, energy is supplied at large scales and cascaded down to smaller and smaller scales. The energy flux can be measured by, e.g., the velocity structure functions. On the other hand, the temporal process of the energy cascade, such as how fast it takes for energy at the forcing scales to transfer down to the dissipative scales, has received relatively little attention. Using novel laboratory turbulent flows and measurement techniques, we experimentally studied the response of turbulence in the inertial and dissipative scales to a sudden excess of energy in the forcing scales. Our measurements give us direct access to the energy cascade process. We also compare our observation with results from direct numerical simulations. [Preview Abstract] |
Tuesday, November 25, 2014 8:39AM - 8:52AM |
M27.00004: Evolution of turbulent kinetic energy in the presence of a uniform kinetic energy gradient without mean shear Adrien Thormann, Charles Meneveau In this work we study grid turbulence with a initial uniform spatial gradient of kinetic energy of the form $k\sim\beta(y-y_0)$, where $y$ is the spanwise position, while having no mean-velocity shear. Therefore, there is no production but only dissipation and spatial transverse diffusion of turbulent kinetic energy. The experiment is performed with the use of an active grid and screens mounted upstream of the wind-tunnel's test section, iteratively designed to produce a uniform gradient of turbulent kinetic energy without mean velocity shear. Data are acquired using X-wire thermal anemometry at different spanwise and downstream locations. Profile measurements are used to quantify the constancy of the mean velocity and the linearity of the initial profile of kinetic energy. Measurements show that at all spanwise locations the decay in the streamwise direction follows a power-law but with exponents $n(y)$ that depend upon the spanwise location. The results are consistent with a parameterization of decay of the form $k/\langle u\rangle^2=\beta (x/x_{\rm ref})^{-n(y)}(y-y_0)/M$. Results for the development of the integral length scale, and for velocity skewness and flatness factors, which show significant deviations from Gaussianity, are also presented. [Preview Abstract] |
Tuesday, November 25, 2014 8:52AM - 9:05AM |
M27.00005: Comparison of fractal and classical grids with the same blockage R. Jason Hearst, Philippe Lavoie Recently, the field of canonical grid turbulence has been reenergized by measurements in the wake of fractal grids. Fractals have produced turbulence that decays more rapidly than traditional grid turbulence experiments. In addition, in the wake of fractals, the normalized dissipation scaling, $C_\epsilon$, appears to grow rapidly, in sharp contrast with traditional expectations that $C_\epsilon \sim$ constant. In the present study, we compare a square-fractal-element grid, composed of a $12 \times 8$ mesh of small square fractal elements, to two regular grids with the same blockage, and similar mesh lengths and thicknesses. The same grid Reynolds number is used so that the results in the wake of the grids are comparable. We also employ a secondary contraction to marginalize anisotropy as a contributing factor to differences in the decay. Ultimately, we demonstrate that in the far-field the turbulence produced by all three grids is similar. However, the development region in the wake of the fractal is extended relative to the classical grids. One of the major conclusions of the present study is that certain classical grid configurations are able to produce higher turbulence intensities and Reynolds number than a fractal for the same blockage. [Preview Abstract] |
Tuesday, November 25, 2014 9:05AM - 9:18AM |
M27.00006: Experimental Study of Homogeneous Isotropic Slowly-Decaying Turbulence in Giant Grid-Wind Tunnel Set Up Alberto Aliseda, Mickael Bourgoin We present preliminary results from a recent grid turbulence experiment conducted at the ONERA wind tunnel in Modane, France. The ESWIRP Collaboration was conceived to probe the smallest scales of a canonical turbulent flow with very high Reynolds numbers. To achieve this, the largest scales of the turbulence need to be extremely big so that, even with the large separation of scales, the smallest scales would be well above the spatial and temporal resolution of the instruments. The ONERA wind tunnel in Modane ($8~m$-diameter test section) was chosen as a limit of the biggest large scales achievable in a laboratory setting. A giant inflatable grid (M=0.8 m) was conceived to induce slowly-decaying homogeneous isotropic turbulence in a large region of the test section, with minimal structural risk. An international team or researchers collected hot wire anemometry, ultrasound anemometry, resonant cantilever anemometry, fast pitot tube anemometry, cold wire thermometry and high-speed particle tracking data of this canonical turbulent flow. While analysis of this large database, which will become publicly available over the next 2 years, has only started, the Taylor-scale Reynolds number is estimated to be between 400 and 800, with Kolmogorov scales as large as a few $mm$. [Preview Abstract] |
Tuesday, November 25, 2014 9:18AM - 9:31AM |
M27.00007: Development and characterization of a Nano-scale temperature probe (T-NSTAP) for turbulent temperature measurement Gilad Arwatz, Yuyang Fan, Carla Bahri, Alexander J. Smits, Marcus Hultmark A new nano-scale temperature probe (T-NSTAP) is presented. The T-NSTAP consists of a miniature, free-standing, platinum wire suspended between silicon supports. The sensor is designed for temperature measurements at high frequencies, operated in constant current mode. The design is based on the cold-wire model proposed by Arwatz et al. (2013) and is shown to have a bandwidth far superior that of conventional cold-wires. This minimizes the effect of temporal filtering as well as spatial filtering on the data and allows for a unique investigation of the full scalar spectrum, including the dissipation range. Data is acquired in a heated grid turbulence setup with constant mean temperature gradient using both the T-NSTAP and a conventional cold wire. It is shown that the cold wire is significantly attenuated over the full range of frequencies including low frequencies with a direct effect on the temperature variance and the scalar rate of dissipation. The model of Arwatz et al (2013) is used to correct the cold wire data and it is shown that the correction works well over the entire spectrum. In addition, the corrected data agrees closely with the T-NSTAP measurements. [Preview Abstract] |
Tuesday, November 25, 2014 9:31AM - 9:44AM |
M27.00008: Scaling of spectra in grid turbulence with a mean cross-stream temperature gradient Carla Bahri, Gilad Arwatz, Michael E. Mueller, William K. George, Marcus Hultmark Scaling of grid turbulence with a constant mean cross-stream temperature gradient is investigated using a combination of theoretical predictions, DNS, and experimental data. Conditions for self-similarity of the governing equations and the scalar spectrum are investigated, which reveals necessary conditions for self-similarity to exist. These conditions provide a theoretical framework for scaling of the temperature spectrum as well as the temperature flux spectrum. One necessary condition, predicted by the theory, is that the characteristic length scale describing the scalar spectrum must vary as $\propto \sqrt{t}$ for a self-similar solution to exist. In order to investigate this, T-NSTAP sensors, specially designed for temperature measurements at high frequencies, were deployed in a heated passive grid turbulence setup together with conventional cold-wires, and complementary DNS calculations were performed to complement and complete the experimental data. These data are used to compare the behavior of different length scales and validate the theoretical predictions. [Preview Abstract] |
Tuesday, November 25, 2014 9:44AM - 9:57AM |
M27.00009: On the Degeneration of Turbulence at High Reynolds Numbers Gregory Bewley, Michael Sinhuber, Eberhard Bodenschatz Turbulent fluctuations in a fluid wind down at a certain rate once stirring has stopped. The role of the most basic parameter in fluid mechanics, the Reynolds number, in setting this decay rate is not generally known. This talk concerns the high-Reynolds-number limit of the process. In a wind-tunnel experiment that reached higher Reynolds numbers than ever before and covered more than two decades in the Reynolds number ($10^4 < Re = UM/\nu < 5\times10^6$), we measured the decay rate with the unprecedented precision of about 2\%. Here $U$ is the mean speed of the flow, $M$ the forcing scale, and $\nu$ the kinematic viscosity of the fluid. We observed that the decay was Reynolds number independent, which contradicts some models and supports others. [Preview Abstract] |
Tuesday, November 25, 2014 9:57AM - 10:10AM |
M27.00010: Time-reversal-symmetry breaking in turbulence Jennifer Jucha, Haitao Xu, Alain Pumir, eberhard Bodenschatz In three-dimensional turbulent flows, the flux of energy from large to small scales breaks time symmetry. We show here that this irreversibility can be quantified by following the relative motion of several Lagrangian tracers. We find by analytical calculation, numerical analysis and experimental observation that the existence of the energy flux implies that, at short times, two particles separate temporally slower forwards than backwards, and the difference between forward and backward dispersion grows as $t^3$. We also find the geometric deformation of material volumes, surrogated by four points spanning an initially regular tetrahedron, to show sensitivity to the time-reversal with an effect growing linearly in $t$. We associate this with the structure of the strain rate in the flow. [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