0

IN THIS ISSUE

TECHNICAL PAPERS

J. Fluids Eng. 2007;129(9):1105-1111. doi:10.1115/1.2754312.

Sheet cavitation has been traditionally analyzed with ideal fluid theory that employs the cavitation number as the single parameter. However, characteristics of cavitation can significantly depend on location of cavity detachment. According to known experimental data, this location is influenced by the freestream speed and the body/hydrofoil size. As shown in this paper, it takes place because of the combined effect of the Reynolds number and Weber number. Here, sheet cavitation is considered as a special kind of viscous separation caused by the cavity itself. The viscous-inviscid interaction concept is employed to analyze the entire flow. Validation of the suggested approach is provided for hydrofoils and bodies of revolution. The effects of flow speed, the body size, and its surface wettability are illustrated by comparison of computed cavity length/shape to the known experimental data. The difference between cavity detachment in laminar and turbulent boundary layers is discussed.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1112-1122. doi:10.1115/1.2754313.

The goal of the study was to explain the relationship between different acoustic signals and visual appearance of cavitation. Measurements of acoustic emission, vibration, and noise were performed on a Kaplan turbine model, with only two blades, in a cavitating condition. Since a model with only two blades was used, most of the side effects were eliminated, and it was concluded that the cavitation itself is the source of the recorded signal. Results showed an interesting relationship between the extent of the cavitation and the recorded data from sensors. At a decreasing cavitation number, the recorded amplitudes from all measurements first rose, experienced a local maximum, then fell to a local minimum, and finally rose again. The cavitation was also visually observed. It was concluded from the measurements that there are distinct correlations between acoustic emission, vibration, and noise on one side and the topology, extent, and type of cavitation structures on the other side. A physical explanation for the phenomenon was introduced and included in a semi-empirical model that links the visual appearance of cavitation on the blade of the turbine to the generated noise and vibration.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1123-1130. doi:10.1115/1.2754326.

The suction performance of turbopumps in cryogenic fluids is basically much better than that in cold water because of the thermodynamic effect of cavitation. However, it is not still clear how the thermodynamic effect works on cavitation instabilities, such as rotating cavitation and cavitation surge. In the present study, the unsteady heat exchange between the cavity and the surrounding liquid is taken into account in a stability analysis using a singularity method. The results are qualitatively compared to existing experiments to clarify the research needs for deeper understanding.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1131-1139. doi:10.1115/1.2754310.

Using electrostatic fields to manipulate and/or pump fluids on the microscale is a promising method for the advancement in microfluidics. Preliminary analysis showed that unidirectional bubble motion could be achieved if the polarization (dielectrophoretic) force could overcome surface tension and viscous forces. Results are presented for the development and fundamental study of dielectrophoretic control of bubble transport in mesochannels. Electrode array configurations were manufactured using printed circuit board technology and mated with an acrylic channel. Bubble velocity, acceleration, and deformation were investigated for a range of bubble sizes, two electrode array configurations, two working fluids—pentane and a 20/80 mixture by mass of ethanol and pentane, two switching frequencies, and a range of $+DC$ pulse applied voltages. A maximum average velocity of $6.6mm∕s$ and a maximum local velocity of $30mm∕s$ were achieved. For the results presented, both the switching frequency and bubble size affected the velocity for a given applied voltage. Of the two fluids tested, there was no measurable difference in the bubble velocity even though the bubble deformation was significantly different for the two fluids. It was concluded that bubble deformation reduced the unidirectional bubble motion effectiveness. Bubble deformation could be reduced by lowering the applied voltage without significantly reducing the velocity of the bubble.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1140-1146. doi:10.1115/1.2754311.

Molecular dynamics (MD) simulations have been performed to provide the basic knowledge of nanofluidics and its applications at the molecular level. A nonequilibrium molecular dynamics (NEMD) code was developed and verified by comparing a micro Poiseuille flow with the classical Navier–Stokes solution with nonslip wall boundary conditions. Liquid argon fluids in a platinum nanotube were simulated to characterize the homogeneous fluid system. Also, positively charged particles were mixed with solvent particles to study the non-Newtonian behavior of the heterogeneous fluid. At equilibration state, the macroscopic parameters were calculated using the statistical calculation. As an application of MD simulation, the nanojetting mechanism was identified by simulating the full process of droplet ejection, breakup, wetting on the surface, and natural drying. For an electrowetting phenomenon, a fluid droplet with positive charges moving on the ultrathin film with negative charges was simulated and then compared to the macroscopic experiments. A conceptual nanopumping system using the electrowetting phenomenon was also simulated to prove its feasibility. The molecular dynamics code developed here showed its potential applicability to the novel concept design of nano- and microelectromechanical systems.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1147-1156. doi:10.1115/1.2754325.

The turbulent flow around square-based, surface-mounted pyramids, of height $h$, in thin and thick boundary layers was experimentally investigated. The influence of apex angle $ζ$ and angle of attack $α$ was ascertained from mean surface flow patterns and ground plane pressure measurements taken at a Reynolds number of $3.3×104$ based on $h$. For both boundary layer flows, it was found that the normalized ground plane pressure distributions in the wakes of all the pyramids for all angles of attack may be scaled using an attachment length $(Xa′)$ measured from the upstream origin of the separated shear layer to the near-wake attachment point on the ground plane. It was also shown that this scaling is applicable to data reported in the literature for other bluff body shapes, namely, cubes, cones, and hemispheres. The ground plane pressure coefficient distributions in the upstream separated flow region, for all the shapes and angles of attack examined, were found to collapse onto two curves by scaling their streamwise location using a length scale $(Xu)$, which is a function of the frontal projected width of the body $(w′)$ and the height of the body. These two curves were for cases where $δ∕h<1$ (“thin” boundary layer) or $δ∕h≥1$ (“thick” boundary layer), where $δ$ is the oncoming boundary layer thickness. Further work is required to provide a more detailed statement on the influence of boundary layer thickness (or state) on the upstream pressure field scaling.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1157-1163. doi:10.1115/1.2754327.

Solute transport in the fractured porous confined aquifer is modeled by the advection-dispersion equation with fractional time derivative of order $γ$, which may vary from 0 to 1. Accounting for diffusion in the surrounding rock mass leads to the introduction of an additional fractional time derivative of order $1∕2$ in the equation for solute transport. The closed-form solutions for concentrations in the aquifer and surrounding rocks are obtained for the arbitrary time-dependent source of contamination located in the inlet of the aquifer. Based on these solutions, different regimes of contamination of the aquifers with different physical properties are modeled and analyzed.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1164-1171. doi:10.1115/1.2754314.

In this study, perforated plates with sharp-edged orificed openings and finite-thickness straight openings were applied to produce nearly isotropic turbulence in a wind tunnel. At the same nominal velocity, the orificed perforated plate was able to produce a higher level of turbulence due to the well-defined flow separation from its sharp edge openings. The integral length, $L$ was found to be related to the square root of the turbulence decay coefficient in the power law decay of turbulence kinetic energy, $A$. The larger $A$ associated with the orificed perforated plate gave rise to a larger $L$. The corresponding streamwise autocorrelation functions for the two perforated plates behaved differently, confirming the quantitative disparity in $L$ and further indicates some qualitative difference in the large-scale structures generated.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1172-1178. doi:10.1115/1.2754318.

The governing equations of two-dimensional steady density currents are solved numerically using a finite volume method. The $v2¯−f$ turbulence model, based on standard $k−ε$ model, is used for the turbulence closure. In this method, all Reynolds stress equations are replaced with both a transport equation for $v2¯$ and an elliptic relaxation equation for $f$, a parameter closely related to the pressure strain redistribution term. The Simple-C procedure is used for pressure-velocity coupling. In addition, Boussinesq’s approximation is used to obtain the momentum equation. The computed height of the progressive density current is compared to the measured data in the literature, resulting in good agreement. The present results show that the flow rate is the most dominant parameter among those affecting the density currents hydrodynamics. The results also show that the $v2¯−f$ turbulence model is able to predict and simulate the characteristics of the low Reynolds turbulent density currents successfully, although it is based on a high Reynolds number turbulence model, i.e., the standard $k−ε$ model. The use of boundary layer convention, saying that the density current’s height is a height at which the concentration is $∼1%$ of the inlet concentration, seems to yield reasonable results.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1179-1185. doi:10.1115/1.2754315.

Turbulent flow structure in a cylinder-on-cone cyclone was experimentally investigated. Measurements were conducted at a fixed geometrical swirl number. Experiments were performed at a swirl number of 3 and Reynolds numbers from 37,100 to 74,200, based on the inlet velocity and the cyclone body diameter. The flow field in planes normal to and through the cyclone axis was measured in detail using a two-component laser Doppler velocimetry (LDA) and a particle imaging velocimetry (PIV). Two dominant frequencies of vortical structures were identified based on LDA-measured tangential and axial velocity spectra. Although one of them agreed quite well with those in literature, the other was reported for the first time. One explanation was proposed.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1186-1192. doi:10.1115/1.2754319.

A logic-based systematic method of designing manifold systems to achieve flowrate uniformity among the channels that interconnect a distribution manifold and a collection manifold has been developed, implemented, and illustrated by case studies. The method is based on tailoring the flow resistance of the individual channels to achieve equal pressure drops for all the channels. The tailoring of the flow resistance is accomplished by the use of gate-valve-like obstructions. The adjustment of the valve-like obstructions is determined here by means of numerical simulations. Although the method is iterative, it may converge in one cycle of the iterations. Progress toward the goal of per channel uniformity can be accelerated by tuning a multiplicative constant. The only departure of the method from being fully automatic is the selection of the aforementioned multiplicative constant. The method is described in detail in a step-by-step manner. These steps are illustrated both generically and specifically for four case studies. As an example, in one of the case studies, an original flow imbalance of over 100% in an untailored manifold system was reduced to a flow imbalance of less than 10% in one cycle of the method.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1193-1202. doi:10.1115/1.2754321.

A commercial computational fluid dynamics (CFD) package was used to develop a three-dimensional, fully turbulent model of the compressible flow across a complex-geometry venturi, such as those typically found in small engine carburetors. The results of the CFD simulations were used to understand the effect of the different obstacles in the flow on the overall discharge coefficient and the static pressure at the tip of the fuel tube. It was found that the obstacles located at the converging nozzle of the venturi do not cause significant pressure losses, while those obstacles that create wakes in the flow, such as the fuel tube and throttle plate, are responsible for most of the pressure losses. This result indicated that an overall discharge coefficient can be used to correct the mass flow rate, while a localized correction factor can be determined from three-dimensional CFD simulations in order to estimate the static pressure at locations of interest within complex venturis.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1203-1211. doi:10.1115/1.2746894.

The flow past a rotating cylinder placed within a uniform stream is investigated at Reynolds numbers ranging from 8500 to 17,000 to 34,000. The dimensionless rotation rate $α$ (ratio of the cylinder peripheral speed to the free-stream velocity) varies from 0 to 7. The experimental investigation is based on laser-Doppler anemometry measurements and particle-image velocimetry (PIV) within a water channel. The analysis of the experimental results mainly concerns the location of the separation points as defined by various criteria. It is found that the criterion suggested by Moore, Rott and Sears (MRS) is met in the case of the downstream-moving walls. Moreover, this study shows that sufficient information was obtained to confirm that the MRS criterion is still valid even in the case of the upstream-moving walls. This is confirmed by the behavior of the vertical velocity component educed from the averaged two-dimensional flow field obtained by PIV measurements.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1212-1227. doi:10.1115/1.2754320.

The purpose of this paper is to present a numerical methodology for the computation of complex 3D turbomachinery flows using advanced multiequation turbulence closures, including full seven-equation Reynolds-stress transport models. The flow equations are discretized on structured multiblock grids, using an upwind biased ($O[ΔxH3]$MUSCL reconstruction) finite-volume scheme. Time integration uses a local dual-time-stepping implicit procedure, with internal subiterations. Computational efficiency is achieved by a specific approximate factorization of the implicit subiterations, designed to minimize the computational cost of the turbulence transport equations. Convergence is still accelerated using a mean-flow-multigrid full-approximation-scheme method, where multigrid is applied only on the mean-flow variables. Speed-ups of a factor 3 are obtained using three levels of multigrid (fine plus two coarser grids). Computational examples are presented using two Reynolds-stress models, and also a baseline $k−ε$ model, for various turbomachinery configurations, and compared to available experimental measurements.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2007;129(9):1228-1240. doi:10.1115/1.2754324.

A multiple surrogate-based optimization strategy in conjunction with an evolutionary algorithm has been employed to optimize the shape of a simplified hydraulic turbine diffuser utilizing three-dimensional Reynolds-averaged Navier–Stokes computational fluid dynamics solutions. Specifically, the diffuser performance is optimized by changing five geometric design variables to maximize the average pressure recovery factor for two inlet boundary conditions with different swirl, corresponding to different operating modes of the hydraulic turbine. Polynomial response surfaces and radial basis neural networks are used as surrogates, while a hybrid formulation of the NSGA-IIa evolutionary algorithm and a $ϵ$-constraint strategy is applied to construct the Pareto front from the two surrogates. The proposed optimization framework drastically reduces the computational load of the problem, compared to solely utilizing an evolutionary algorithm. For the present problem, the radial basis neural networks are more accurate near the Pareto front while the response surface performs better in regions away from it. By using a local resampling updating scheme the fidelity of both surrogates is improved, especially near the Pareto front. The optimal design yields larger wall angles, nonaxisymmetrical shapes, and delay in wall separation, resulting in 14.4% and 8.9% improvement, respectively, for the two inlet boundary conditions.

Commentary by Dr. Valentin Fuster

TECHNICAL BRIEF

J. Fluids Eng. 2007;129(9):1241-1244. doi:10.1115/1.2771578.

A potential flow and viscous flow solver have been coupled to produce a robust computational tool useful for the design of low-speed wind tunnel contractions. After validation against published numerical and experimental wind tunnel data, the method is used to evaluate recently proposed contraction shapes from the literature. The results show that, on balance, a fifth-order polynomial provides a good design solution. Newly proposed shapes will either improve available flow area at the expense of contraction outlet flow uniformity or vice versa.

Commentary by Dr. Valentin Fuster