Ram Balachandar

Rough Wall Turbulent Boundary Layers in Shallow Open Channel Flow

An experimental study was undertaken to investigate the effects of roughness on the structure of turbulent boundary layers in open channels. The study was carried out using a laser Doppler anemometer in shallow flows for three different types of rough surface, as well as a hydraulically smooth surface. The flow Reynolds number based on the boundary layer momentum thickness ranged from 1400 to 4000. The boundary layer thickness was comparable with the depth of flow and the turbulence intensity in the channel flow varied from 2 to 4 percent. The defect profile was correlated using an approach which allowed both the skin friction and wake strength to vary. The wake parameter was observed to vary significantly with the type of surface roughness in contradiction to the wall similarity hypothesis. Wall roughness also led to higher turbulence levels in the outer region of the boundary layer. The profound effect of surface roughness on the outer region as well as the effect of channel turbulence on the main flow indicates a strong interaction, which must be accounted for in turbulence models

Variable threshold outlier identification in PIV data

This paper describes a variable threshold technique that can be applied to any particle image velocimetry (PIV) post-analysis outlier identification algorithm which uses a threshold such as the local median or the cellular neural network techniques. Although these techniques have been shown to work quite well with constant thresholds, the selection of the threshold is not always clear when working with real data. Moreover, if a small threshold is selected, a very large number of valid vectors can be mistakenly rejected. Although careful monitoring may alleviate this danger in many cases, that is not always practical when large data sets are being analysed and there is significant variability in the properties of the vector fields. The method described in this paper adjusts the threshold by calculating a mean variation between a candidate vector and its eight neighbours. The main benefit is that much smaller thresholds can be used without suffering catastrophic loss of valid vectors. The main challenge in obtaining this threshold field is that it must be based on a filtered field to be representative of the underlying velocity field. In this work, a simple median filter which requires no threshold was used for preliminary rejection. A local threshold was then calculated from the mean difference between each vector and its neighbours. The threshold field was also filtered with a Gaussian kernel before use. The algorithm was tested and compared to the base techniques by generating artificial velocity fields with known numbers of spurious vectors. For these tests, the ability of the algorithms to identify bad vectors and preserve good vectors was monitored. In addition, the technique was tested on real PIV data from the developing region of an axisymmetric jet. The variable threshold versions of these algorithms were found to be much less susceptible to erroneously rejecting good vectors. This is because the variable threshold techniques extract information about the local velocity gradient from the data themselves. The user-adjustable parameters for the variable threshold methods were found to be more universal than the constant threshold methods.

Revisiting turbulence in smooth uniform open channel flow

This study reexamines the mean velocity scaling as well as higher order turbulent moments in a uniform smooth open channel flow with three different aspect ratios. In the overlap region, the velocity profiles follow the classical logarithmic law. In the outer region, the mean velocity data at various aspect ratios collapse on to each other only when the length scale is suitably modified. This length scale is defined on the basis of a region of constant turbulence intensity close to the free surface and is equal to the depth of flow at large aspect ratios. The proposed new length scaling also provides for a positive value of the wake parameter. Furthermore, turbulence distributions including that of the Reynolds shear stress, and triple correlations and collapse onto a single line making them nearly independent of aspect ratio. Quadrant decomposition of the velocity data was used to quantify the differences in the turbulence structure at the three channel aspect ratios. The quadrant analysis shows that the turbulence in open channel is similar to that in both turbulent boundary layers and flow in two-dimensional channels when all turbulent events are included. When only the extreme events are considered, differences between open channel flow and turbulent boundary layers become significant. The conditional quadrant analysis reveals that the violent ejections do penetrate into the flow and they are responsible for producing large portion of the Reynolds shear stress. Some effects of aspect ratio are revealed when the ratio of the ejection to sweep events are calculated. The turbulent events with the higher aspect ratios tend to be closer to the two-dimensional channel data.

Turbulent Boundary Layers in Low Reynolds Number Shallow Open Channel Flows

The results of an experimental investigation of turbulent boundary layers in shallow open channel flows at low Reynolds numbers are presented. The study was aimed at extending the database toward lower values of Reynolds number. The data presented are primarily concerned with the longitudinal mean velocity, turbulent-velocity fluctuations, boundary layer shape parameter and skin friction coefficient for Reynolds numbers based on the momentum thickness (Reθ ) ranging from 180 to 480. In this range, the results of the present investigation in shallow open channel flows indicate a lack of dependence of the von Karman constant κ on Reynolds number. The extent to which the mean velocity data overlaps with the log-law decreases with decreasing Reθ . The variation of the strength of the wake with Reθ is different from the trend proposed earlier by Coles.

Investigation of Fluid Structures in a Smooth Open-Channel Flow Using Proper Orthogonal Decomposition

This paper reports particle image velocimetry (PIV) measurements of the instantaneous velocity fields in a smooth open-channel flow. The Reynolds number of the flow based on the water depth was 21,000. The instantaneous velocity fields were analyzed using proper orthogonal decomposition (POD) to expose the vortical structures. The velocity fields were reconstructed using different combination of modes; the first 12 modes to expose the energetic structures, and from Modes 13 to 100 to expose the less energetic structures. The first set recovered about 50% of the turbulent kinetic energy while the second group of modes recovered about 33% of the energy. The POD results were further combined with the results from the momentum analysis as well as with the conditional quadrant analysis performed at three different threshold levels. The POD results revealed the existence of hairpin vortices of different sizes and energy levels. Most of the large eddies are elongated and inclined toward the boundaries in the str...

Quantitative investigation of vortical structures in the near-exit region of an axisymmetric turbulent jet

High-resolution particle image velocimetry (PIV) measurements were made at the exit of an axisymmetric free jet at a Reynolds number of 21 900 with a top-hat, low-turbulence exit velocity profile. The data were analysed using the proper orthogonal decomposition (POD) technique to identify the main energy-containing vortices. The vortices so identified have been further quantified by computing their size, position, circulation and direction of rotation. The technique employed resolved vortices of radius 0.055 jet exit diameters and larger. This quantitative information has also been related to observations from flow visualisation images and measurements of the downstream evolution of the jet. The data clearly show the formation of alternating direction toroidal vortices identified in the fluctuating fields which begin to be resolved at a streamwise location of one-half the jet exit diameter. The vortices grow in number approximately linearly in the downstream direction until about one jet diameter. After this, the number of vortices decreases and their size and circulation increase as they move downstream. This is indicative of the vortex pairing process. The high spread rate of the current jet is attributed to the vigorous production of large-scale vortical structures in this near-exit region due to the very near top-hat exit velocity profile and the resulting strong shear layer near the exit.

Characteristics of Shallow Turbulent Near Wakes at Low Reynolds Numbers

The present study deals with the noninvasive measurement of both velocity and concentration in the near region of shallow turbulent wakes using a laser-Doppler anemometer and a video-imaging technique. A 40 mm wide flat plate placed normal to the flow is used as the wake generator. The flow depths considered in the present study are small compared to the width of the channel and the generated wakes are categorized as shallow. Tests were conducted at two depths of flow (h =20 and 40 mm) and the boundary layer thickness of the approaching flow is comparable to the depth. The Reynolds number of the flow based on the approaching freestream boundary layer momentum thickness varies from 180 to 400, while, the Reynolds number based on the test body width was maintained nearly constant (4000). Measurements were carried out at three axial stations (2.5, 5, and 10 plate widths) downstream of the bluff body. At each axial station, the velocity measurements were carried at distances of h/4, h/2, and 3h/4 from the channel bottom and spanning the cross section of the wake. Appropriate length and velocity scales are identified to characterize the wake

Numerical simulation of high-speed turbulent water jets in air

Numerical simulation of high-speed turbulent water jets in air and its validation with experimental data has not been reported in the literature. It is therefore aimed to simulate the physics of these high-speed water jets and compare the results with the existing experimental works. High-speed water jets diffuse in the surrounding atmosphere by the processes of mass and momentum transfer. Air is entrained into the jet stream and the entire process contributes to jet spreading and subsequent pressure decay. Hence the physical problem is in the category of multiphase flows, for which mass and momentum transfer is to be determined to simulate the problem. Using the Eulerian multiphase and the k–ε turbulence models, plus a novel numerical model for mass and momentum transfer, the simulation was achieved. The results reasonably predict the flow physics of high-speed water jets in air.

Analysis of Coherent Structures in the Far-Field Region of an Axisymmetric Free Jet Identified Using Particle Image Velocimetry and Proper Orthogonal Decomposition

This paper investigates an isothermal free water jet discharging horizontally from a circular nozzle (9 mm) into a stationary body of water. The jet exit velocity was 2.5 m/s and the exit Reynolds number was 22,500. The large-scale structures in the far field were investigated by performing a proper orthogonal decomposition (POD) analysis of the velocity field obtained using a particle image velocimetry system. The number of modes used for the POD reconstruction of the velocity fields was selected to recover 40% of the turbulent kinetic energy. A vortex identification algorithm was then employed to quantify the size, circulation, and direction of rotation of the exposed vortices. A statistical analysis of the distribution of number, size, and strength of the identified vortices was carried out to explore the characteristics of the coherent structures. The results clearly reveal that a substantial number of vortical structures of both rotational directions exist in the far-field region of the jet. The number of vortices decreases in the axial direction, while their size increases. The mean circulation magnitude is preserved in the axial direction. The results also indicate that the circulation magnitude is directly proportional to the square of the vortex radius and the constant of proportionality is a function of the axial location.

Open channel boundary layer relaxation behind a forward facing step at low Reynolds numbers

We report laser-Doppler anemometer measurements of mean velocity and turbulence statistics upstream and downstream of a 3-mm forward facing step in a shallow open channel flow. The Reynolds numbers based on the momentum thickness (θ) of the approaching upstream flow and step height (h) are in the range 1010≤Re θ ≤2240 and 960≤Re h ≤1890, respectively. Measurements are obtained at 50 step heights upstream of the leading edge of the step and 1≤x/h≤162 downstream of the step