0


TECHNICAL PAPERS

J. Fluids Eng. 2003;125(3):401-413. doi:10.1115/1.1568355.

Partition of the stress dissipation has been studied in an axisymmetric strained flow field to assess the possible existence of local isotropy for turbulence at small scales. This is a simple flow to study because the axes of anisotropy of the Reynolds stresses and of the dissipation tensor are aligned. Using invariant theory, the relationship between the stress and dissipation tensors was derived, satisfying restrictions for the limiting states of turbulence and the assumed behavior for large Reynolds number and small anisotropy. The role of the anisotropy in constraining models for the turbulent dissipation rate and the pressure-strain correlations is discussed. Comparisons of the resulting closure with experimental data for several axisymmetric flows are good within the limitations of the data.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):414-427. doi:10.1115/1.1567306.

The use of dense gases in many technological fields requires modern fluid dynamic solvers capable of treating the thermodynamic regions where the ideal gas approximation does not apply. Moreover, in some high molecular fluids, nonclassical fluid dynamic effects appearing in those regions could be exploited to obtain more efficient processes. This work presents the procedures for obtaining nonconventional thermodynamic properties needed by up to date computer flow solvers. Complex equations of state for pure fluids and mixtures are treated. Validation of sound speed estimates and calculations of the fundamental derivative of gas dynamics Γ are shown for several fluids and particularly for Siloxanes, a class of fluids that can be used as working media in high-temperature organic Rankine cycles. Some of these fluids have negative Γ regions if thermodynamic properties are calculated with the implemented modified Peng-Robinson thermodynamic model. Results of flow simulations of one-dimensional channel and two-dimensional turbine cascades will be presented in upcoming publications.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):428-437. doi:10.1115/1.1566047.

The spatial stability of a natural convection flow on upward-facing, heated, inclined plates is revisited. The eigenvalue problem is solved numerically employing two methods: the collocation method with Chebyshev polynomials and the fourth-order Runge-Kutta method. Two modes, traveling waves and stationary longitudinal vortices, are considered. Previous theoretical models indicated that nonparallel effects of the mean flow are significant for the vortex instability mode, but most of them ignored the fact that the eigenfunctions are dependent on the streamwise coordinate as well. In the present work, the method of multiple scales is applied to take the nonparallel flow effects into consideration. The results demonstrate the stabilizing character of the nonparallel flow effects. The vortex instability mode is also considered within the scope of partial differential equations. The results demonstrate dependence of the neutral point on the initial conditions but, farther downstream, the results collapse onto one curve. The marching method is compared with the quasi-parallel normal mode analysis and with theoretical results including correction to nonparallel flow effects. The marching method provides better agreement of theoretical and experimental growth rates.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):438-442. doi:10.1115/1.1567311.

The problem of separated flow in bends of arbitrary turning angles has been examined. The method of analysis is based on the inviscid flow theory coupled with Kirchhoff’s separation model. The physical flow problem is first transformed to the hodograph domain, and then into a rectangular computational region using properly selected flow parameters. The solution is first established in the hodograph plane. The final flow pattern including the inner and outer walls of the bend, the separation streamline, and other flow properties in the physical plane are subsequently obtained through direct integration. The results of the present analysis are compared with those of Lichtarowicz and Markland as well as Mankbadi and Zaki.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):443-446. doi:10.1115/1.1567309.

The viscous flow in a curved tube with partial slip on the boundary occurs in many practical situations. The problem is formulated in curved tube coordinates and solved by perturbation for small curvature. The mutual interaction of slip, curvature, and inertia causes changes in the axial flow, surface shear, and secondary flow. It is found that the net flow increases with increased slip and decreased Reynolds numbers.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):447-458. doi:10.1115/1.1566048.

A sensitivity analysis is done for turbulent cavitating flows using a pressure-based Navier-Stokes solver coupled with a phase volume fraction transport model and nonequilibrium k-ε turbulence closure. Four modeling parameters are assessed, namely, Cε1 and Cε2, which directly influence the production and destruction of the dissipation of turbulence kinetic energy, and Cdest and Cprod, which regulate the evaporation and condensation of the phases. Response surface methodology along with design of experiments is used for the sensitivity studies. The difference between the computational and experimental results is used to judge the model fidelity. Under noncavitating conditions, the best selections of Cε1 and Cε2 exhibit a linear combination with multiple optima. Using this information, cavitating flows around an axisymmetric geometry with a hemispherical fore-body and the NACA66(MOD) foil section are assessed. Analysis of the cavitating model has identified favorable combinations of Cdest and Cprod. The selected model parameters are found to work well for different geometries with different cavitation numbers for attached cavity. It is also confirmed that the cavitation model parameters employed in the literature are within the range identified in the present study.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):459-468. doi:10.1115/1.1568358.

For high-speed cavitating flows, compressibility becomes significant in the liquid phase as well as in the vapor phase. In addition, the compressible energy equation is required for studying the effects of the propulsive jet on the cavity. Therefore, a numerical method is developed to compute cavitating flows over high-speed torpedoes using the full unsteady compressible Navier-Stokes equations. The multiphase system of equations is preconditioned for low-speed flow computations. Using the mass fraction form, we derive an eigensystem for both the conditioned and the nonconditioned system of equations. This eigensystem provides stability for the numerical discretization of the convective flux and increases the convergence rate. This method can be used to compute single as well as multiphase flows. The governing equations are discretized on a structured grid using an upwind flux difference scheme with flux limits. Single as well as multiphase flows are computed over a cavitating torpedo. The results indicate that the preconditioned system of equations converges rapidly to the required solution at very low speeds. The theoretical results are in good agreement with the measurements.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):469-478. doi:10.1115/1.1567308.

Measurements of local void fraction, rise velocity, and bubble diameter have been obtained for cocurrent, wall-heated, upward bubbly flows in a pressurized refrigerant. The instrumentation used are the gamma densitometer and the hot-film anemometer. Departure bubble size is correlated in terms of liquid subcooling and bulk bubble size in terms of void fraction. Flow visualization techniques have also been used to understand the two-phase flow structure and the behavior of the bubbly flow for different bubble shapes and sizes, and to obtain the bubble diameter and rise velocity. The lift model is provided explicitly in terms of Eotvos number which is changed by changing the system pressure. In general, Eotvos number plays a strong role in determining both bubbly lift and drag. Such insight coupled with quantitative local and averaged data on void fraction and bubble size at different pressures has aided in developing bubbly flow models applicable to heated two-phase flows at high pressure.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):479-485. doi:10.1115/1.1566049.

A new technique of measuring void fraction in magnetic fluid using electromagnetic induction was proposed. In order to establish the measuring method, a feasibility study was conducted experimentally with an aid of numerical analysis. From the results of static experiment and numerical analysis, it was obtained that there exists a linear relationship between the void fraction and the measured electromotive force, when induction coils were connected in series for Helmholtz excitation coils, regardless of distribution of air bubbles in magnetic fluid. By applying the calibrated linear relationship to actual two-phase situations, it was revealed that the proposed method yielded quite reasonable account for measuring the void fraction, showing excellent agreement with the mechanical measured data in the two-phase flow apparatus, and with the published correlation of the drift flux model. From the results of the present investigation, it was proved that the proposed technique is feasible for the actual measurement of void fraction in two-phase flow of magnetic fluid.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):486-499. doi:10.1115/1.1566046.

Here we report on results obtained from large eddy simulations of flow inside a stirred tank performed using a spectral multidomain technique. The computations were driven by specifying the impeller-induced flow at the blade tip radius. Stereoscopic PIV measurements along with a theoretical model are used in defining the impeller-induced flow as a superposition of circumferential, jet and tip-vortex pair components. Both time-independent (fixed inflow) and time-dependent (oscillatory inflow) impeller-induced flows were considered. In both cases, the improved impeller-induced inflow allowed for the development of tip-vortex pairs in the interior of the tank. At Rem=4000 considered here, the flow in the interior of the tank naturally evolves to a time-dependent turbulent state. The jet component of the impeller-induced flow becomes unstable and shows signs of both sinuous and varicose behavior. The vortex pairs are anchored near the blades, but as they extend outwards into the tank their backbones exhibit time-dependent fluctuation. The instability of the jet is intimately connected with the fluctuation of the tip-vortex system. The time-averaged location of the vortex backbone compares well with previous measurements. The radial profile of θ-averaged radial velocity along the midplane is a good sensitive measure for evaluating the computed results. It is observed that computed flow from the 20 deg oscillatory impeller-induced inflow model compares well with the corresponding experimental measurements on the r-z plane.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):500-509. doi:10.1115/1.1567305.

Flow around a simplified bus is analyzed using large-eddy simulation. At the Reynolds number of 0.21×106, based on the model height and the incoming velocity, the flow produces features and aerodynamic forces relevant for the higher (interesting in engineering) Reynolds number. A detailed survey of both instantaneous and time-averaged flows is made and a comparison with previous knowledge on similar flows is presented. Besides the coherent structures observed in experimental and previous numerical studies, new smaller-scale structures were registered here. The mechanisms of formation of flow structures are explained and the difference between instantaneous and time-averaged flow features found in the experimental observations is confirmed. Aerodynamic forces are computed and their time history is used to reveal the characteristic frequencies of the flow motion around the body. A comparison is made of pressure and velocity results with experimental data and shows fairly good agreement.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):510-519. doi:10.1115/1.1567310.

While a lot of attention was given to shock wave reflections from wedges during the past four decades, only little work was published regarding the similar case of blast wave reflection from wedges. In the present paper this subject is studied experimentally and theoretically/numerically. The obtained results show that the geometry of the reflected wave pattern is similar in the two cases when both incident waves have the same initial pressure jump across their fronts. However, different reflected pressure signatures (history) are observed in these two cases. The pressures obtained behind a reflected shock wave are always higher than those obtained behind the corresponding similar blast wave. In the present case differences as high as 17% were observed.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):520-532. doi:10.1115/1.1570859.

Gasoline direct injection (GDI) spark ignition engines may be able to run over a wide range of operating conditions. The GDI process allows combustion with lean mixtures which may lead to improved fuel economy and emissions relative to homogeneous spark ignition (SI) engines. To satisfy the different modes of operation, the tuning of GDI engines requires a large number of engine tests which are time-consuming and very expensive. To reduce the number of tests, a model with a very short computational time to simulate the engines in the whole operating range is needed; therefore the objective of this paper is to present a reduced model to analyze the combustion process in GDI engines, applied to a homogeneous stoichiometric mode. The objective of the model is to reproduce the same tendencies as those obtained by three-dimensional models, but with a reduced computational time. The one-dimensional model is obtained thanks to a reduction methodology based on the geometry of the combustion front computed with three-dimensional models of the KIVA-GSM code, a modified version of KIVA-II code including a CFM combustion model. The model is a set of n one-dimensional equations (i.e., for n rays), taking into account a thin flame front, described with the flamelet assumption. It includes a CFM combustion model and a (k,ε)-model including the mean air motions (swirl and tumble). The results of the one-dimensional model are compared to those obtained by the KIVA IIGSM under different engine conditions. The comparison shows that the one-dimensional model overestimates the maximum cylinder pressure, which has an insignificant effect on the net indicated work per cycle. The results obtained by the numerical simulations are close to those given by the three-dimensional model, with a much reduced computation time.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):533-542. doi:10.1115/1.1570860.

The unsteady turbulent inflow into a swirl-inducing stator upstream of propeller (SISUP) propeller is presented. The upstream stators and hull boundary layer generate a complex, three-dimensional inflow that was measured using x-wire anemometry. High resolution measurements consisting of 12 locations in the radial direction and 600 in the circumferential direction yielded mean velocity and rms turbulent quantities for a total of 7200 points. The axial, radial, and circumferential velocity fields were thus measured. This enabled the induced velocity due to the stator wakes, the induced velocity due to the propeller, and the turbulent hull boundary layer to be characterized. To assist in decoupling the effects on the velocity field due to the stator and propeller, a potential flow computation of the swirl component was used. Spectra and autocorrelation analysis of the inflow velocity field were used to estimate the integral length scale and lend further insight into the turbulent flow structure. These data can be used to validate computational fluid dynamics codes and assist in developing of turbulent inflow models.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):543-550. doi:10.1115/1.1568356.

The unsteady viscous flow fields of a cross-flow fan are computed by time-accurately solving the two-dimensional incompressible Navier-Stokes equations with the unstructured triangular mesh solver algorithms. Based on pressure fluctuation data acquired at the surfaces of 35 rotating blades and stabilizer, acoustic pressures are predicted by the Ffowcs Williams-Hawkings equation. The aerodynamic noise sources of the cross-flow fan are also identified by correlating the acoustic pressure fluctuations with the unsteady flow characteristics during one revolution of the impeller. The present method is applied to the uniform and random pitch fans to investigate their performance and aeroacoustic noise characteristics, especially the frequency modulation of the tonal noise at the blade passing frequency (BPF).

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):551-560. doi:10.1115/1.1568360.

Interference effects on vortex-induced vibrations of two side-by-side elastic cylinders, fixed at both ends (with no deflection and displacement) in a cross-flow, were experimentally investigated. The dynamic responses of the cylinders were measured using two fiber-optic Bragg grating (FBG) sensors. Simultaneously, a single hot wire was used to measure the velocity in the wake. It has been previously observed that violent resonance occurs when transverse cylinder spacing ratio, T/d, is either large (>2.0) or small (<1.2), but not for intermediate cylinder spacing, i.e., T/d=1.2∼2.0. This work aims to improve the understanding of the physics behind this observation, and mostly focuses on the fluid-structure interaction in the flow regime of intermediate cylinder spacing. It is well known that in this flow regime the fluid dynamics around one cylinder is totally different from that around the other; the vortical structures are characterized by different dominant frequencies, i.e., about 0.1 and 0.3 (normalized), respectively. The present data indicates that the vortical structures at these frequencies are either weak or different in the formation process from the case of T/d>2.0 or T/d<1.2, thus resulting in a weak excitation and subsequently an absence of violent resonance. The interrelationship between the vortical structures generated by the two cylinders is also investigated and interpreted in terms of different vortex generation mechanisms. The different fluid dynamics around each cylinder is further found to be responsible for a deviation between the natural frequencies of the combined fluid-cylinder system associated with each cylinder.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):561-568. doi:10.1115/1.1568357.

High-pass filtering of instantaneous two-dimensional PIV data is employed to educe vortices occurring in the axial plane of a self-similar turbulent axisymmetric jet. An automated method is used to identify the vortices and to measure circulation, tangential velocity, vorticity, and centrifugal force within their cores (defined here as the region within the largest closed streamline). Results include radial variation of these quantities within vortex cores, and the energy of vortices. We find that the vorticity is maximum at the vortex center and decreases monotonically with the radial coordinate. Results indicate that the center of a larger vortex spins faster than a smaller vortex (in an ensemble averaged sense); however, the trend reverses to give the expected result at the core edge. Vorticity results for different vortex radii collapse upon normalization. The average energy of vortices is seen to increase as the square of the vortex radius. In addition, three possible regimes of vortex number versus vortex size are suggested by our data.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):569-575. doi:10.1115/1.1568359.

Experimental measurements of the mean wall shear stress and boundary layer momentum thickness on long, thin cylindrical bodies are presented. To date, the spatial growth of the boundary layer and the related boundary layer parameters have not been measured for cases where δ/a (a=cylinder radius) is much greater than one. Moderate Reynolds numbers (104<Reθ<105) encountered in hydrodynamic applications are considered. Tow tests of cylinders with diameters of 0.61, 0.89, and 2.5 mm and lengths ranging from approximately 30 meters to 150 meters were performed. The total drag (axial force) was measured at tow speeds up to 17.4 m/sec. These data were used to determine the tangential drag coefficients on each test specimen, which were found to be two to three times greater than the values for the corresponding hypothetical flat-plate cases. Using the drag measurements, the turbulent boundary layer momentum thickness at the downstream end of the cylindrical bodies is determined, using a control volume analysis. The results show that for the smallest diameter cylinders, there is no indication of relaminarization, and a fully developed turbulent boundary layer exists. A scaling law for the momentum thickness versus length Reynolds number is determined from the data. The results indicate that the spatial growth of the boundary layers over the entire length is less than for a comparable flat-plate case.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):576-585. doi:10.1115/1.1567307.

The effect of inverter ripple current on fuel cell stack performance and stack lifetime remains uncertain. This paper provides a first attempt to examine the impact of inverter load dynamics on the fuel cell. Since reactant utilization is known to impact the mechanical nature of a fuel cell, it is suggested that the varying reactant conditions surrounding the cell govern, at least in part, the lifetime of the cells. This paper investigates these conditions through the use of a dynamic model for the bulk conditions within the stack, as well as a one-dimensional model for the detailed mass transport occurring within the electrode of a cell. These two independent modeling approaches are used to verify their respective numerical procedures. In this work, the inverter load is imposed as a boundary condition to the models. Results show the transient behavior of the reactant concentrations within the stack, and of the mass diffusion within the electrode under inverter loads with frequencies between 30 Hz and 1250 Hz.

Commentary by Dr. Valentin Fuster

TECHNICAL BRIEFS

J. Fluids Eng. 2003;125(3):586-589. doi:10.1115/1.1566042.

The CFD performance estimation of turbo booster vacuum pump shows the axial vortex and back flow is evident when the mass flow rate is increased. The pressure is increased from the pump inlet to the outlet for the low mass flow rate cases. But for high mass flow rate cases, the pressure is increased until the region near the end of the rotor then decreased. The calculated inlet pressure, compression ratio, and pumping speed is increased, decreased, and decreased, respectively, when the mass flow rate is increased. The pumping speed is increased when the rotor speed is increased.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 2003;125(3):590-594. doi:10.1115/1.1566043.

Korotky,  G. J., and Taslim,  M. E., 1998, “ Rib Heat Transfer Coefficient Measurements in a Rib-Roughened Square Passage,” ASME J. Turbomach., JOTUEI120, pp. 376–385.97rJOTUEI0889-504XTaslim,  M. E., and Lengkong,  A., 1998, “ 45° Staggered Rib Heat Transfer Coefficient Measurements in a Square Channel,” ASME J. Turbomach., JOTUEI120, pp. 571–580.97rJOTUEI0889-504XHwang,  J. J., and Liou,  T. M., 1994, “ Augmented Heat Transfer in a Rectangular Channel With Permeable Ribs Mounted on the Wall,” ASME J. Turbomach., JOTUEI116, pp. 912–920.97rJOTUEI0889-504XMolki,  M., Faghri,  M., and Ozbay,  O., 1994, “ A New Correlation for Pressure Drop in Array of Rectangular Blocks in Air-Cooled Electronic Units,” ASME J. Fluids Eng., JFEGA4116, pp. 856–861.97kJFEGA40098-2202Hong,  J. C., and Hsieh,  S. S., 1993, “ Heat Transfer and Friction Factor Measurements in Ducts With Staggered and In-Line Ribs,” ASME J. Heat Transfer, JHTRAO115, pp. 58–65.97lJHTRAO0022-1481Greiner,  M., Chen,  R. F., and Wirtz,  R. A., 1991, “ Enhanced Heat Transfer and Pressure Drop Measured From a Flat Surface in a Grooved Channel,” ASME J. Heat Transfer, JHTRAO113, pp. 498–501.97lJHTRAO0022-1481Garimella,  S. V., and Schlitz,  D. J., 1992, “ Influence of Fin Aspect Ratio on Heat Transfer Enhancement,” SAE Journal of Materials and Manufacturing,ZZZZZZ101, pp. 467–472.Sparrow,  E. M., and Hajiloo,  A., 1980, “ Measurements of Heat Transfer and Pressure Drop for an Array of Staggered Plates Aligned Parallel to an Air Flow,” ASME J. Heat Transfer, JHTRAO102, pp. 426–432.97lJHTRAO0022-1481Cur,  N., and Sparrow,  E. M., 1978, “ Experiments on Heat Transfer and Pressure Drop for a Pair of Collinear, Interrupted Plates Aligned With the Flow,” Int. J. Heat Mass Transf., IJHMAK21, pp. 1069–1080.ijhIJHMAK0017-9310Halle,  H., Chenoweth,  J. M., and Wambsganss,  M. W., 1988, “ Shellside Waterflow Pressure Drop Distribution Measurements in an Industrial-Sized Test Heat Exchanger,” ASME J. Heat Transfer, JHTRAO110, pp. 60–67.97lJHTRAO0022-1481Kim,  J. Y., Lai,  M. C., Li,  P., and Chui,  G. K., 1995, “ Flow Distribution and Pressure Drop in Diffuser-Monolith Flows,” ASME J. Fluids Eng., JFEGA4117, pp. 362–368.97kJFEGA40098-2202ASME, 1959, “Fluid Meters—Their Theory and Applications,” Report of ASME Research Committee on Fluid Meters, New York.Moffat,  R. J., 1985, “ Using Uncertainty Analysis in the Planning of an Experiment,” ASME J. Fluids Eng., JFEGA4107, pp. 173–185.97kJFEGA40098-2202

J. Fluids Eng. 2003;125(3):595-596. doi:10.1115/1.1566044.

The oxide formation on the surface of the molten metal jet was shown to have a drastic effect on the droplet formation process according to the description of some publication. Thus, the main objective of this research is to investigate the influence of oxygen concentration on the breakup and the monosized droplets generation of molten metal jet (Sn63 Pb37 alloy). The breakup phenomena of molten metal jet can be approximately divided into three regimes. They are “breakup regime” for oxygen concentration below C1, “transition regime” for oxgyen concentration between C1 and C2, and “breakup failing regime” for oxygen concentration beyond C2, respectively.

J. Fluids Eng. 2003;125(3):597-599. doi:10.1115/1.1566045.

A two-dimensional air jet, heated at a density ratio of 0.8, under external forcing by flexible wires is investigated experimentally. In each shear layer of the hot jet flow, a wire of diameter 0.23 mm (0.015 jet width) is flexibly mounted along the spanwise direction. By flow visualization, temperature measurements, and spectral analysis, the study demonstrates that the wires have quite different effects on the jet flow depending on that the wires are motionless or vibrating in the flow, and the shear layers of the heated plane jet can be manipulated by means of flexible wires.

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In