0

Fluids Machinery Forum

2005;():1-9. doi:10.1115/FEDSM2005-77003.

The design and performance of an inlet throttle mechanism composed of a spool and a direct acting solenoid is demonstrated to yield similar fuel injection characteristics, described by injection traces, as a pressure regulator placed in the output of a HEUI (hydraulic electronic fuel injector) pump with the added advantage of yielding torque savings of approximately 50% through the speed and load operation map in a typical automotive application. The paper details: the throttle mechanism, a variable gain area with a linear schedule designed for fine flow control followed by a high-gain to remedy cold start; the solenoid design, optimized to yield a constant force over the entire stroke for a given current level; and the close-loop feedback algorithm, characterized by a feed-forward term determined from bench testing at discrete pump speeds and loads. Testing shows the flow desired to collapse onto a single band correlating with duty cycle to the solenoid, a robust feature that enables the feed-forward term of the control system to be effective in adjusting to transient conditions. The paper includes comparisons at 1200, 2000 and 3000 pump rpm of the inlet throttle mechanism and of a pressure regulator applied to a HEUI pump in a system that comprises the pump, high-pressure accumulator and 8 fuel injectors. Time traces of pressure in the accumulator and rates of injected fuel at the cylinders indicate the inlet throttle device is slower to recover pressure following an injection event as expected, but only at lower operating speeds, which is due to the limited response of the pressure regulator at the higher speeds. The rates of injection traces show that there is no deterioration of the fuel injection characteristics.

Commentary by Dr. Valentin Fuster
2005;():11-15. doi:10.1115/FEDSM2005-77018.

In this investigation, two new 3D wicket gates are designed for improving Francis turbine’s stability under high water head and low discharge condition. Numerical simulation and experiment are applied for studying influence of wicket gate type. The model Francis-turbine with 3D wicket gate is computed on three dimensional turbulence method. Base on standard k -ε turbulence model and SIMPLEC algorithm, the whole water-passage from inlet of case to outlet of draft is computed together, then distribution of velocity and pressure can be obtained. Based on experiment and computation result, comparison of performance and vorticity character between different wicket gates is made.

Commentary by Dr. Valentin Fuster
2005;():17-25. doi:10.1115/FEDSM2005-77057.

In hydraulic systems, flow control valve is used to regulate the flow of fluid to actuators by adjusting the valve opening. However, the inlet and the outlet pressures of the valve are not always remaining constant. Any change in pressure will alter the flow rate through the valve and alter the actuator speed consequently. Pressure compensated flow control valves are often used in hydraulic systems when accurate speed control is required under varying supply or load pressures. The basic structure of the pressure compensated flow control valve is by incorporating a compensating spool to maintain a constant pressure drop across the metering orifice. Under ideal circumstance, the actuator speed can be constant and controllable, regardless of load or system pressure changes. However, in practical applications, any system or load pressures variations will cause force unbalance on valve compensating spool and affect the control accuracy. The steady and dynamic response of the flow control valve plays an important role on hydraulic system behavior. Therefore, analyzing and understanding of the valve steady and dynamic behaviors is very important. In this study, the steady and dynamic performance of a pressure compensated flow valve is simulated numerically by solving the characteristic equations. The parameters studied in this research are biased spring constant, pre-compressed spring length, spool mass, and the damping orifice characteristics. The simulation results show that the flow force is identified as the key factor to affect the control accuracy. Increasing the spring constant as well as the pre-compressed spring length will increase the steady flow rate and reduce the transient response time. Decreasing the damping orifice opening or the discharge coefficient will increase the transient response time. The spool mass has practically no effect on the flow rate.

Commentary by Dr. Valentin Fuster
2005;():27-34. doi:10.1115/FEDSM2005-77086.

The stability of a pressure compensating circuit used to regulate the output pressure in a variable displacement radial piston pump is considered here. A throttle spool placed in the pump inlet controls the flow into the pump to match the flow requirement of the system. A pilot, solenoid operated, with a hydraulic main stage, pressure-regulating valve in the pump outlet regulates the system pressure. The discharge flow of the pressure regulator controls the inlet throttle spool position. Too large of a flow spill through the pressure regulator denoting excess pump flow results in flow and pressure build up in the inlet throttle control cavity so as to throttle the inlet flow. Hydraulic instability may arise if the control pressures applied to the inlet throttle device over restrict the inlet flow. Stability is jeopardized when the inlet and outlet flows are too closely matched. Excessive pressure build-up in the actuating volume acting over the inlet throttle piston yields a phase-shift between the piston motion and the system pressure response. This phase-shift clearly affects the stable operation of the control system Tightening the reaction between components in the system requires limiting the pressure acting over the inlet throttle control piston. Lower pressures are obtained by opening up the bleed orifice in the control circuit. Other design elements including the inlet throttle flow area vs. stroke schedule, main stage area gain and pilot geometry are considered. The advantages of an overland over a poppet design in the main stage are illustrated in the correction of a rolling instability. Experimental tests monitored torque, pressures in both the pilot and main stages of the valve, pressure at the pump outlet and in the system accumulator.

Commentary by Dr. Valentin Fuster
2005;():35-41. doi:10.1115/FEDSM2005-77294.

In the previous paper, the optimum meridian profile of impeller was obtained for various specific speed by means of eight shape factors, that is, inlet relative flow angle β1 , turning angle Δβ, axial velocity ratio kc = Cm2 /Cm1 , impeller diameter ratio kd = D1c /D2c , outlet hub-tip ratio ν2 , tip solidity σtimp , mid span solidity σcimp and hub solidity σhimp . In this paper, the optimum meridian profile of multi-stage impeller with guidevane was obtained by means of twelve shape factors. The additional four shape factors are guidevane tip solidity σtgv , mid span solidity σcgv , hub solidity σhgv and coefficient of peripheral velocity at impeller inlet or guidevane outlet kCu1c . In the optimum method, the hydraulic efficiency and suction specific speed are calculated by diffusion factor. In the optimum condition, the best hydraulic efficiency or the best suction specific speed is obtained. In the cyclic flow condition of multi-stage impeller with guidevane, the absolute flow velocity of guidevane outlet is equal to that of impeller inlet, and the diameter of guidevane outlet is equal to that of impeller inlet. In this calculation, the diameter of impeller outlet is equal to that of guidevane inlet. The total calculation number of case study is very large, so the number of each parameter is about between four and seven. The best 1000 optimum meridian profiles and the best design parameter are selected for few kinds of specific speed using twelve dimensional optimum method. As the result of this calculation, the optimum meridian profile of multi-stage impeller and guidevane. The more detailed optimum multi-stage mixed flow impeller and guidevane profile is drawn. For, example, the 1000 specific speed is selected for case study of multi-stage mixed flow impeller. At first, the approximate optimum shape factors are present shape factors. And the optimum shape factors which have better efficiency are tried to find near the present shape factors. Then the study of shape factor changes is the objective of this paper.

Topics: Impellers
Commentary by Dr. Valentin Fuster
2005;():43-45. doi:10.1115/FEDSM2005-77322.

Flow field of centrifugal pump (Pump flow passage includes suction pipe, impeller and volute casing.) obtained by solving the Reynolds average Navier-Stokes equation. Computational performance curve is compared with measuring curve. The result demonstrates that the computational head versus discharge curve is coincident with the measuring curve; the computational efficiency curve is coincident with the measuring curve in large-capacity area, in lower-capacity area not so well. Though the simulation result is not accurate highly in part operating domains, the pressure and velocity distribution and other flow characteristics in whole flow path in all operating domain produced by the simulation are also useful to analyze flow problems in dredge centrifugal pump.

Commentary by Dr. Valentin Fuster
2005;():47-51. doi:10.1115/FEDSM2005-77363.

In the past twenty years wind energy remerged on the world scene with a very healthy growth rate, it has outstripped photovoltaics as the world’s fastest growing energy source, with a growth rate in excess of 30 percent per annum. The proposed paper presents a numerical procedure for the analysis and design of Horizontal Axis Wind Turbine rotors for fabrication in countries with limited manufacturing base and limited design expertise. To ascertain the accuracy and to determine where further improvements could be initiated; numerical findings were then compared with published experimental test data, the compression showed an average deviation of less than 3%. Once the approach was validated and standardized an airfoil design was produced. A computational fluid dynamic code coupled with a simple numerical algorithm aided the inverse design procedure. The final design is well proportioned and theoretically able to meet the stated objective and satisfied the constraints. The generated geometrical data is in a form suitable for manufacture using local manufacturing capabilities, the primary objective of this work.

Commentary by Dr. Valentin Fuster
2005;():53-57. doi:10.1115/FEDSM2005-77385.

In this study, an optimization design method is established for a rotor blade of a Curtis turbine. Bezier curve is generally used to define the profile of turbine blades. However, this curve is not proper to a supersonic impulse turbine. Section shape of a supersonic turbine blade is composed of straight lines and circular arcs. That is, it has several constraints to define the section shape. Thus, in this study, a blade design method is developed by using B-spline curve in which local control is possible. The turbine blade section has been changed by varying three design parameters of exit blade angle, stagger angle and maximum camber. Then flow analyses have been carried out for the sections. Lift-drag ratio of the blade section is used as the object function, and it is maximized in the optimization. Second-order response surface model is employed to express the object function as a function of design parameters. Central composite design method is used to reduce the number of design points. Then, an evolution strategy is employed to obtain the optimized section of the Curtis turbine blade.

Commentary by Dr. Valentin Fuster
2005;():59-64. doi:10.1115/FEDSM2005-77388.

A tandem arrangement of double rotating cascades and single diffuser cascade, proposed as a centrifugal pump with high performance in air-water two-phase flow condition, yields lower head due to the smallness of the impeller outlet in comparison with a impeller with large outlet diameter and no diffuser. Influences of impeller diameter change and installation of diffuser blades on two-phase flow performance were experimentally investigated under the case of the same volute casing. As the result, the similarity law of the diameter of impeller having the similar blade geometry and the rotational speed is satisfied even in two-phase flow condition. Comparing pump performances between a large impeller without diffuser blades and a small one with diffuser blades, higher two-phase flow performance is obtained by controlling the rotational speed of a small impeller with diffuser blades in the range of small water flow rates, while a large impeller with no diffuser gives high performance in the range of high water flow rate and small air flow rate.

Commentary by Dr. Valentin Fuster
2005;():65-69. doi:10.1115/FEDSM2005-77389.

We designed an axial fan for servers using computational fluid dynamics (CFD) and numerical optimization. The performance of the fan, namely static pressure rise and efficiency, was calculated using commercial CFD software based on an incompressible Reynolds-averaged Navier-Stokes (RANS) solver. An automatic program developed in-house was used to generate the grids for CFD calculation. Numerical optimization—using a simulated annealing algorithm (SA)—was used for determining the optimized shape of the fan. After optimizing the fan, initial and optimized fan designs were made for experiments using rapid prototyping, and their performances, based on such things as efficiency and noise level, were measured. Results demonstrated that the optimized fan design achieved higher efficiency than the initial design. Multi optimization was also developed for maximizing the fan efficiency and minimizing the casing height. An additional finding was that there was a trade-off between the fan efficiency and casing height.

Topics: Design , Fans , Optimization
Commentary by Dr. Valentin Fuster
2005;():71-76. doi:10.1115/FEDSM2005-77419.

The knowledge of the flow behavior in pump stages which consist of an impeller, a bladed diffuser and a bladed return channel is of great importance for the design of multistage centrifugal pumps. Especially the Interaction of the impeller flow with the stationary diffuser blades and the behavior of the return channel blades affect the efficiency of a pump stage in a considerable way. In this contribution the transient flow in an industrial centrifugal pump stage, which has an impeller with seven blades, a radial diffuser with ten blades and a return channel with also ten separate blades, has been simulated numerical by using the commercial software code CFX-5.7. Because of the unfavorable ratio of blade numbers a complete meshing of all flow channels was necessary. In consequence the cumulative amount of grid nodes reached a number of nearly 6 million nodes. As a result of the numerical investigation of the time dependent flow accomplished for this contribution, the influence of the rotating impeller on the flow in the stationary parts of the pump is presented in detail. All flow parameters are shown as a function of time and are discussed with respect to the position of the impeller relative to the stator blades.

Commentary by Dr. Valentin Fuster
2005;():77-88. doi:10.1115/FEDSM2005-77426.

All the physical parameters, such as energy, force, velocity, and acceleration are constructed from two different kinds; one is real and the other is imaginary. Their acting directions are normal to each other. The former acts horizontal direction and causes visible kinetic movement on fluid particle. All the supplied energy is utilized and consumed. The latter acts vertical direction but does not cause any visible kinetic movement on fluid particle. All the energy transfer from mechanical to hydraulic and from hydraulic to mechanical is caused by the imaginary parameters in vertical direction.

Commentary by Dr. Valentin Fuster
2005;():89-94. doi:10.1115/FEDSM2005-77436.

In this work, an experimental study about the influence of some geometric features on the aeroacoustic behavior of a squirrel-cage fan, used in automotive air conditioning units, has been carried out. The study focused on the effect of both the shape and the position of the volute tongue on the noise generated by the fan. Different geometric configurations were tested in order to compare the results. First of all, the performance curves were measured in a standardized test facility. Then, the acoustic behavior of the fan was characterized by means of acoustic pressure measurements near the fan inlet. The comparison of the test results indicated a great dependence of both the shape and the position of the volute tongue and the noise generation. In particular, some geometric configurations of the volute tongue were able to reduce the fan noise generation without reducing the fan performance.

Commentary by Dr. Valentin Fuster
2005;():95-98. doi:10.1115/FEDSM2005-77440.

This paper will explain the numerical analysis and the mapping of the flow in a confined region. In this paper some characteristics of non-steady, compressible, flow are explored, including compression and expansion wave interactions and creation. The results will show a promising achievement, first, to understand the flow structure inside a supersonic confined region, second, to use this knowledge to interpolate the numerical results in order to achieve a design methodology that will benefit the industrial applications for example in turbomachinery. Results including contour plots of static pressure, total pressure, and Mach number will show the structure of oblique shock waves in a complex three-dimensional conical surface. A CFD analysis enables one to understand the complex flow structure inside this confined region. Through this computational analysis, a better interpretation of the physical phenomenon of the three dimensional rotting oblique shock waves can be achieved. It is essential to evaluate the ability of numerical technique that can solve problems in which compression and expansion waves occur. In particular it is necessary to understand the details of developing a mesh that will allow resolution of some discontinuities in similar flow.

Topics: Shock waves , Rotors
Commentary by Dr. Valentin Fuster
2005;():99-106. doi:10.1115/FEDSM2005-77447.

The safe and reliable operation of industrial facilities and high pressure test stands for engine and component testing is largely dependent on the smooth performance of control valves. However, such valves frequently experience pressure oscillations from hydrodynamic instabilities, cavitation and unsteady valve operation. In this paper, we present a series of high fidelity computational simulations of control valves primarily to understand the physics associated with the dominant instability modes. A generalized multi-element framework with sub-models for grid adaption, grid movement and multi-phase flow dynamics was used to carry out the simulations. We discuss the methodology in detail with the example of transient analyses of a gaseous hydrogen control valve and capture the fluid dynamic instability that results from valve operation. Additionally, we provide detailed analyses of a modal instability that is observed in the operation of a pressure regulator valve. In both cases, the instabilities are not localized and manifest themselves as a system wide phenomena leading to oscillations in mass flow and/or undesirable chatter.

Commentary by Dr. Valentin Fuster
2005;():107-111. doi:10.1115/FEDSM2005-77452.

This paper investigates the influence of labyrinth seal teeth damage due to rotor impacting on the performance and the rotordynamic characteristics of impeller eye seals in centrifugal compressors. A well-established CFD-perturbation model was employed to predict the rotordynamic coefficients. The inclusion of at least an approximate shroud leakage path chamber is prefered for accurate prediction of seal-inlet swirl velocity and flow-induced rotordynamic forces. Impeller eye seals with teeth damage: (a) suffer significant leakage increases due to the increased seal clearance and (b) produce higher seal-inlet swirl velocity as well as larger rotordynamic forces, which tend to cause the system to become unstable. It was also found that distorted teeth tip geometries have an insignificant influence on both leakage and rotordynamic coefficients. The leakage path influence on seal-inlet swirl velocity W0 was also explored to thoroughly understand the rotordynamic characteristics of the eye seal subject to various degrees of teeth damage.

Commentary by Dr. Valentin Fuster
2005;():113-118. doi:10.1115/FEDSM2005-77453.

Rotors in steam turbines experience significant axial shifting during start-up and shut-down process due to thermal expansion. This axial-shifting could significantly alter the flow pattern and the flow-induced rotordynamic forces in labyrinth seals, which in turn, can considerably affect the rotor-seal system’s performance. This paper investigates the influence of the rotor-axial-shifting on leakage rate and rotordynamic forces for high-low labyrinth seals under different geometrical and operational conditions. A well-established CFD-perturbation model was employed to predict the rotordynamic coefficients. A surprisingly large effect was found for rotordynamic characteristics due to changes in seal configurations caused by rotor axial shifting. It was also found that less destabilizing effect arose from rotor-axial-shifting in the leakage flow direction whereas a more destabilizing effect arose from shifting against the leakage flow direction. A tentative explanation was proposed for the large sensitivities of dynamic forces to the off-design operations with rotor-axial-shifting.

Commentary by Dr. Valentin Fuster
2005;():119-124. doi:10.1115/FEDSM2005-77455.

Unavoidable rotordynamic impacting on labyrinth seal teeth sometimes occurs when centrifugal compressors, for example, undergo transients. Consequently, the labyrinth seal teeth are damaged or disfigured in various ways when the surface opposite to the teeth is non-abradable. Thus far, no quantitative information concerning the effect on seal leakage is available. The present work focuses on the effect of seal leakage due to such permanently bent labyrinth seal teeth. The investigation was done numerically by solving the 2-D, axisymmetric RANS equations with a finite-volume algorithm. The high-Reynolds number k-ε turbulence model was used with standard wall functions. A broad variety of tooth seal bending was studied by varying the bending curvature and the length of bending, as well as the after-bend tooth radial clearance. The results show that the bending damage drastically affects the leakage as well as the flow pattern. This is due largely to the altered clearance caused by the bending. However, other bending factors, such as the bending curvature and the percentage of tooth length that is bent, also contribute to the change of leakage and flow pattern.

Topics: Leakage
Commentary by Dr. Valentin Fuster
2005;():125-131. doi:10.1115/FEDSM2005-77460.

A pump with a specific speed of 12000 was chosen to operate as a turbine (PAT) for a micro-hydro site having 5 m of head. Turbine performance of the pump was unavailable so it was simulated using CFD. The CFD model was first verified by comparison of simulated pump performance and manufacturer data. Simulated PAT performance covered a range of flow rates, from one to three times that of pump best efficiency point (BEP), for blade angles of 0 and ± 4°. The PAT BEP was located at a flow rate of 1.4 times that of pump BEP and a head of 1.6 times. For the specific site this corresponded to a shaft power of 32 k W and a flow rate of 770 1/s. The PAT was found to have an extended range of good efficiency, > 60%, for up to 3 times the pump BEP flow rate.

Commentary by Dr. Valentin Fuster

Forum on Unsteady Flows

2005;():133-144. doi:10.1115/FEDSM2005-77097.

Pulsatile flow patterns in an intracranial side-wall aneurysm of a human carotid artery were investigated experimentally using microscopic particle image velocimetry on a flexible wall model. Numerical calculations were performed on a stiff-walled artery with the same geometry and flow conditions. Experimental results showed cyclical vortex formation and decay within the aneurysm. Numerical simulations agreed well with the experimental results and indicate that vortex position and magnitude are due to pulsatile flow conditions rather than wall elasticity. However, the present results suggest that wall compliance plays a role in sustaining the pressure gradients needed for vortex development.

Commentary by Dr. Valentin Fuster
2005;():145-153. doi:10.1115/FEDSM2005-77220.

Flow fields of two dimensional jets impinging on the sharp edge are computationally simulated and the effect of various parameters on the edgetone that is created by the flow interaction is investigated. Compressible Navier-Stokes equations are used so that acoustic waves are captured accurately as a part of feedback-loop. For numerical accuracy, Pade type compact finite difference scheme are used. First parameter is the jet velocity. Computational result shows good qualitative agreement with the experiment. Edgetone frequencies obtained by the computation also show good correspondence with those of experimental study in the past. Second parameter is the nozzle lip thickness. Although not considered in the computational study in the past, the nozzle lip thickness influences to the results. Amplitude of acoustics of larger nozzle lip is greater than that of smaller ones. This effect may comes from the fact that acoustic wave as a part of feedback loop is emphasized by nozzle lip. Third parameter is the jet-profile. Four different jet-profiles with the same maximum velocity (from top-hat profile to parabolic profile) and four different jet-profiles with the same mean velocity are computed. The mean jet velocity appears to have strong influence on the stage. The results also indicated that the mean jet velocity and the jet-profile have influence on edgetone frequencies.

Commentary by Dr. Valentin Fuster
2005;():155-160. doi:10.1115/FEDSM2005-77236.

Experimental investigations of the flow around an elliptic cylinder of axis ratio 1:3 were carried out for several angles of attack in a wide range of Reynolds number. The flow characteristics were studied by measuring the fluid forces and the surface pressure. In the critical Reynolds number regime, a discontinuous change of flow state was observed. This change was accompanied by the remarkable hysteresis phenomenon. Details of this hysteresis process are described in the paper.

Commentary by Dr. Valentin Fuster
2005;():161-168. doi:10.1115/FEDSM2005-77335.

Present paper is devoted to numerical investigation of unsteadiness caused by impeller-diffuser interaction in a 8:1 total pressure ratio centrifugal compressor. The compressor designed by CIAM [7], and manufactured and tested by Customer gave satisfactory performances even under the first test. Further development requires new insights and advanced numerical tools. In this context, this paper presents Navier-Stokes computations of 3D viscous unsteady flow field within the impeller-diffuser configuration. Steady and unsteady computations indicated spacious zone of low velocity / reverse flow on pressure surface of the diffuser vane. To suppress this reverse flow, new vaned diffuser has been tailored through application of 3D inverse design procedure for Navier-Stokes equations [8]. Subsequent steady and unsteady N-S calculations performed for compressor with the new diffuser demonstrated depression of reverse flow within diffuser and different unsteady loading of the diffuser vane.

Commentary by Dr. Valentin Fuster
2005;():169-174. doi:10.1115/FEDSM2005-77422.

A numerical study is conducted to investigate steady and pulsed fluidic actuation in transonic flow over an open cavity. Numerical results are obtained for the unsteady three-dimensional flow with three different steady mass injection rates and one pulsed injection upstream of the cavity. The simulations are carried out using the full 3-D Navier Stokes equations with the two-equation k-ε based Detached Eddy Simulation (DES) model to calculate the flow and acoustic fields. Computational results are presented for unsteady pressure fluctuations, vorticity contours and kinetic energy profiles at different injection ratios. The sound pressure level (SPL) and the kinetic energy spectra highlight the effectiveness of actuation in tone attenuation at peak frequencies. The computed sound pressure level (SPL) spectra with and without injection are compared with available experimental data and LES predictions.

Commentary by Dr. Valentin Fuster
2005;():175-180. doi:10.1115/FEDSM2005-77431.

The present paper deals with numerical investigations of vortex shedding from a circular cylinder in uniform and shear flows. The computational approach is based on solving the complete Transient Reynolds-Averaged Navier-Stokes (TRANS) equations. The ζ-f model proposed recently by Hanjalic et al. (2004) is used for comparison with the recent computations of Basara et al. (2003) who used the Reynolds-stress model (RSM) developed by Speziale et al. (1991) and the hybrid EVM/RSM turbulence model developed by Basara and Jakirlic (2003). The recent study of the same flows (Basara, Jakirlic and Alajbegovic (2003)) showed that the wall treatment has an important role in correct predictions of forces acting on a cylinder. In this work, a new compound wall treatment (CWT) of Popovac and Hanjalic (2005) was used. A simple modification to the treatment of the dissipation rate of the original CWT is proposed. Furthermore, a new νt - k - ζ - f model is modeled on the basis of an exact form derived from the ζ-f model and the first results are given here. Calculations with the ζ-f model employed as a low-Re number model are performed and the results are compared with those obtained with the same model in conjunction with CWT.

Commentary by Dr. Valentin Fuster
2005;():181-186. doi:10.1115/FEDSM2005-77438.

Liquid-liquid emulsions will undergo a phase inversion, in which the dispersed phase becomes the continuous phase and vice versa, under certain conditions. A phase inversion is not a smooth transition and an emulsion close to the inversion point may oscillate back and forth between the oil-in-water (O/W) and water-in-oil (W/O) forms, creating flow instabilities that may be detrimental in certain industrial situations. The results of laboratory laminar flow experiments in which an aqueous and an organic liquid phase are emulsified as they flow through a circular tube containing a commercial high-shear static mixer are discussed. As the concentration of the dispersed phase and/or its flow rate is increased, flow instabilities are initiated in the test-section and are measured as fluctuations in pressure drop. The intensity of these fluctuations reaches a maximum as the liquid-liquid system approaches the phase inversion point. Once a stable phase inversion is achieved, the fluctuations subside. This phenomenon was observed over a wide range of viscosity ratios for the two liquid phases, but was absent for low viscosity ratios and low-shear static mixers.

Topics: Laminar flow , Water
Commentary by Dr. Valentin Fuster
2005;():187-190. doi:10.1115/FEDSM2005-77442.

This paper will explain the numerical analysis and the mapping of the flow in a confined region. In this paper some characteristics of non-steady, compressible, flow are explored, including compression and expansion wave interactions and creation. The results will show a promising achievement, first, to understand the flow structure inside a supersonic confined region, second, to use this knowledge to interpolate the numerical results in order to achieve a design methodology that will benefit the industrial applications for example in turbomachinery. Results including contour plots of static pressure, total pressure, Mach number, temperature and velocity vectors will show the structure of rotating oblique shock waves in a complex three-dimensional conical surface. A CFD analysis enables one to understand the complex flow structure inside this confined region. Through this computational analysis, a better interpretation of the physical phenomenon of the three dimensional rotting oblique shock waves can be achieved. It is essential to evaluate the ability of numerical technique that can solve problems in which compression and expansion waves occur. In particular it is necessary to understand the details of developing a mesh that will allow resolution of some discontinuities in similar flow.

Commentary by Dr. Valentin Fuster
2005;():191-198. doi:10.1115/FEDSM2005-77446.

The biologically-inspired method of trailing-edge articulation is investigated as a means of reducing tonal noise due to the stator wake / rotor blade interaction in underwater vehicles. This work is experimental in nature and conducted in the closed channel water tunnel at Naval Undersea Warfare Center in Newport, Rhode Island. Tail articulation is carried out with a life scale stator model with hinged flapping tail designed to (i) operate in freestream velocities corresponding to Reynolds number in the range 75,000 < Re < 300,000 and (ii) operate at frequencies up to 30 Hz in order to investigate the range of Strouhal number 0.0 < St < 0.35. Velocity measurements in the active stator wake are carried out by Laser Doppler Velocimetry (LDV) and Particle Image Velocimetry (PIV) in order to investigate the effects of tail articulation on the stator wake. Time averaged measurements of the stator wake by LDV show that Strouhal number of the tail articulation has a dominant effect on the time mean stator drag. Instantaneous phase-averaged measurements of the stator wake by PIV show three regimes of the stator wake as Strouhal number is increased; quasi-steady wake spreading, vortex roll up, and strong vortex wake. Ongoing experiments with an instrumented propeller will demonstrate the efficacy of stator trailing-edge articulation on reducing unsteady blade forces.

Commentary by Dr. Valentin Fuster
2005;():199-206. doi:10.1115/FEDSM2005-77458.

Separating oscillating flow in an internal adverse pressure gradient geometry is studied experimentally. Phase-locked PIV measurements and simultaneous pressure measurements reveal that during the accelerating portion of the cycle, the flow remains attached in spite of a very large adverse pressure gradient. During the decelerating portion of the cycle, the flow is more prone to separation. The duration and extent of the separation depend strongly on the oscillation displacement amplitude relative to the cross-stream dimension. In some cases, the flow separates but reattaches as the separated shear layer is accelerated temporally. The time-varying pressure measurements are used to determine the resultant minor losses for the flow in each direction. These are found to be an increasing function of displacement amplitude and independent of the Reynolds number.

Commentary by Dr. Valentin Fuster
2005;():207-212. doi:10.1115/FEDSM2005-77473.

The detailed wake structure behind pitching airfoil and heaving airfoil at a low Reynolds number region was measured by PIV. Moreover, dynamic thrust acting on them in water tunnel was measured by a six-axes sensor. At the high non-dimensional trailing edge velocity and the non-dimensional heaving velocity, the thrust producing vortex street is formed clearly. Moreover, it has been founded that not only the distance between vortices becomes narrow but also vorticity increases as the non-dimensional trailing edge velocity and the non-dimensional heaving velocity increase. The averaged dynamic thrust acting on a pitching airfoil and a heaving airfoil increases as the non-dimensional trailing edge velocity and the non-dimensional heaving velocity increase. The hysteresis loops of dynamic thrust acting on a pitching airfoil and a heaving airfoil show reentrant and convexity shapes characteristics. The dynamic behavior of dynamic thrust acting on a heaving airfoil is different from that on a pitching airfoil. The thrust efficiency of a pitching airfoil increased up to Vp = 0.7 rapidly and maximum thrust efficiency was 0.34. The thrust efficiency of a heaving airfoil increased up to Vp = 0.5 rapidly and the maximum thrust efficiency was 0.20.

Topics: Thrust , Wakes , Airfoils
Commentary by Dr. Valentin Fuster

4th Forum on CFD Applications in Large Facilities

2005;():213-222. doi:10.1115/FEDSM2005-77027.

The blast furnace is a huge counter-flow heat exchanger, lined with refractory brick, which is used to chemically reduce and physically convert iron oxides into liquid iron called “hot metal”. In the blast furnace, iron ore, coke and limestone are dumped into the top, and preheated air is blown into the bottom. The raw materials require 6 to 8 hours to descend to the bottom of the furnace, called the hearth, where they become the final product of liquid slag and liquid iron. The liquid products are drained (tapped) from the hearth at regular intervals. The hearth is a crucial region of the blast furnace, since the life of its refractory determines to a great extend the life span of the furnace. Therefore, it is necessary to understand the process of erosion and the wear mechanisms. The wear of the hearth lining (refractory) arises from a combination of hydrodynamic, chemical, and thermomechanical phenomena. Due to the very hostile environment inside the furnace, direct measurements of flow and temperature distributions that impact hearth wear is fundamentally precluded. Consequently, one resorts to physical and mathematical modeling of the process. In the current study, a mathematical model of the tapping process is presented. This model, in the current stage of an ongoing program in McMaster University, is two-dimensional and unsteady. The coke bed (packed bed or deadman) is assumed of uniform permeability. The effect of the coke-free layer height on the flow pattern and bottom wall shear stress distributions is investigated. Also, the effect of the taphole height is considered in other wards, the effect of the sump ratio is studied. The study is performed using Fluent which is a commercial computational fluid dynamics software package. From the study it was shown that, for a sitting bed, the flow resistance is uniform every where and liquid flows directly to the taphole along the shortest path which offers the least resistance in this case. When the packed bed (deadman) floats at a low height the liquid now has a region with much less resistance to flow in. Therefore, the liquid rushes into the coke-free layer putting higher stress on the hearth floor which means higher heat transfer rates and more erosion of the bottom wall of the hearth. The higher the floating is the weaker this effect. Changing the sump ratio also affects the stress load on the hearth floor. Deeper pool, higher taphole, has a less shear stress on the floor compared to a shallow one, lower taphole.

Commentary by Dr. Valentin Fuster

Forum on Prediction of the Aero-Acoustics Noise Generated by Turbomachines

2005;():223-226. doi:10.1115/FEDSM2005-77103.

Put abstract text here. A serial of experiments were conducted to study the noise radiated from a series connected nozzle pair. The experiment results are presented in this paper. This nozzle pair consists of two nozzles, one is called source nozzle, and the other is a secondary nozzle. In these experiments, the structure of source nozzle was fixed while that of secondary nozzle was changeable. The source nozzle is mounted on a pressure chamber which is connected to an air compressor. A steel tube is fixed at the tail of source nozzle. The secondary nozzle is connected to the other end of the tube. Throat size of secondary nozzle is larger then that of source nozzle. 15 types of nozzles with different expansion ratio, length of expand segment, and throat structure were used as the secondary nozzle. Jet noise pressure of these nozzle pairs was measured by 40AF Free Field Microphone. The frequency spectrum of jet noise from source nozzle with steel tube under different chamber pressures was calculated. The pressure range is from 0.1 to 1.2 MPa. This result is compared with those spectrums of nozzle pair with different secondary nozzle under different chamber pressures. The trend of peak frequency shifts for different nozzle pair and different chamber pressure is presented in this paper. The secondary nozzles make frequency peak shift from the source nozzle, especially in low frequency band. Different structure of secondary nozzle has different influence on the frequency characteristics of jet noise. Length of expand segment has greater influence on low frequency peak than other two factors. Joint time-frequency analysis is also used in analyze the change of frequency spectrum during throat size decreased under fixed chamber pressure and various spectrograms are also presented. In low frequency band, frequency peak remains during the change of source nozzle throat size. But in higher frequency band, the frequency peak shifts from low frequency to higher ones as the throat size decreases.

Commentary by Dr. Valentin Fuster
2005;():227-236. doi:10.1115/FEDSM2005-77134.

High speed vaned centrifugal fans are widely used in several manufacturing and home appliances. For the designers the noise generated by these machines is one of the most important parameters to be reduced. The centrifugal fan used for this study is made up of an impeller, a diffuser and a return channel. The impeller turns at a relatively high speed about 35000 rpm. The objective of this study is to understand the mechanism of the noise generation within this type of machines. The contribution of the tangential and radial forces is highlighted. These fluctuating forces are due to the unsteady flow at the impeller-diffuser interface. The obtained result shows the effect of monopole and dipole sources on the overall noise.

Commentary by Dr. Valentin Fuster
2005;():237-246. doi:10.1115/FEDSM2005-77424.

The purpose of this study is to determine the effect of the sweep angle used in axial flow fans on their aerodynamic and acoustic behavior. To do so, two fans, having the same aerodynamic characteristics, were designed with forward and radial sweep angles. The 3D numerical simulation allowed obtaining steady and unsteady loading on the blades. Instantaneous velocity profiles, located downstream of the fans, will be presented and compared to the experimental data. The Ffowcs Williams & Hawkings formulation was used to model the acoustic spectra. The results from CFD simulation (acoustic rotating dipole) were used as input data for the acoustic modeling. Predicted and measured acoustic spectra will be presented. It was found that the forward fan has a uniform radial distribution of kinetic turbulence. Also, the three downstream velocity components have better radial distributions. This result was confirmed by the experimental and predicted aeroacoustic results.

Topics: Fans , Axial flow
Commentary by Dr. Valentin Fuster

Forum on Advances in Free Surface and Interface Fluid Dynamics

2005;():247-252. doi:10.1115/FEDSM2005-77274.

The continuum surface force (CSF) method has been extensively employed in the volume-of-fluid (VOF), level set (LS) and front tracking methods to model surface tension force. It is a robust method requiring relatively easy implementation. However, it is known to generate spurious currents near the interface that may lead to disastrous interface instabilities and failures of grid convergence. A different surface tension implementation algorithm, referred to as the pressure boundary method (PBM), is introduced in this study. The surface tension force is incorporated into the Navier-Stokes equation via a capillary pressure gradient while the free surface is tracked by a coupled level set and volume-of-fluid (CLSVOF) method. It has been shown that the spurious currents are greatly reduced by the present method with the sharp pressure boundary condition preserved. The numerical results of several cases have been compared with data reported in the literature and are found to be in a close agreement.

Commentary by Dr. Valentin Fuster
2005;():253-258. doi:10.1115/FEDSM2005-77275.

The time-dependent relaxation dynamics of a moderately elongated liquid ligament has been studied numerically. The Navier-Stokes equations are solved using a finite-volume formulation with a two-step projection method on a fixed grid. The free surface of the liquid ligament is tracked by a coupled level set and volume-of-fluid (CLSVOF) method with the surface tension force determined by the continuum surface force (CSF) model. The relaxation process of a free elongated liquid ligament has been simulated and the numerical results are in agreement with findings reported in the literature. The end-pinching mechanism of the breakup process has been thoroughly examined. The determining factor for reopening of a pinching neck has been identified. The effects of several parameters on the relaxation mechanism have also been examined. It has been found that the initial end shape of the ligament and the Ohnesorge number play a vital role in the overall relaxation process.

Commentary by Dr. Valentin Fuster
2005;():259-264. doi:10.1115/FEDSM2005-77367.

For interface-tracking simulation of two-phase flows, we propose a new computational method, NS-PFM, combining Navier-Stokes (NS) equations with phase-field model (PFM). Based on the free energy theory, PFM describes an interface as a volumetric zone across which physical properties vary continuously. Surface tension is defined as an excessive free energy per unit area induced by density gradient. Consequently, PFM simplifies the interface-tracking procedure by use of a standard technique. The proposed NS-PFM was applied to several problems of incompressible, isothermal two-phase flow with the same density ratio as that of an air-water system. In this method, the Cahn-Hilliard (CH) equation was used for predicting interface configuration. It was confirmed through numerical simulations that (1) the flux driven by chemical potential gradient in the CH equation plays an important role in interfacial advection and reconstruction, (2) the NS-PFM gives good predictions for pressure increase inside a bubble caused by the surface tension, (3) coalescence of liquid film and single drop falling through a stagnant gas was well simulated, and (4) collapse of liquid column under gravity was predicted in good agreement with other available data. Then, another version of NS-PFM was proposed and applied to a direct simulation of bubble nucleation of a non-ideal fluid in the vicinity of the critical point, which demonstrated the capability of NS-PFM to capture liquid-vapor interface motion in boiling and condensation.

Commentary by Dr. Valentin Fuster
2005;():265-270. doi:10.1115/FEDSM2005-77396.

A volume tracking scheme, ASCA (Advanced Subgrid Counting Algorithm), which is easily extendable to three-dimensions and possesses a good volume conservation property, is proposed. To examine the potential of ASCA, several two-phase flow simulations are carried out. As a result, the following conclusions are obtained: (1) Predicted shapes and breakup characteristics of single fluid particles in simple shear flows agree well with available data even with a low spatial resolution, (2) A water drop impinging on water surface and an air bubble rising through a stagnant water are successfully simulated with little errors in volume conservation, (3) Predicted shapes and terminal velocities of single drops in stagnant liquids under wide ranges of the viscosity ratio and Morton number agree well with measured one, and (4) In spite of a low spatial resolution, vapor bubbles flowing in a strong shear flow are simulated with good volume conservation. The interface sharpness is well preserved even after a large deformation of bubbles.

Topics: Algorithms
Commentary by Dr. Valentin Fuster
2005;():271-279. doi:10.1115/FEDSM2005-77404.

Experimental studies were made on the gas-core turbulence structure in vertical upward annular two-phase flow passing through a round tube with a throat section. The gas-core turbulence is affected and modified by the dynamic interaction between gas-core flow and liquid film flow through the wavy interface. The test channel has a throat section, which consist of nozzle, throat and diffuser part, where the cross sectional area of the channel is changed along the flow direction. In the throat section, the flow is in transient state because the flow is accelerated or decelerated along the flow direction as the cross sectional area of the channel is changes. The gas-phase turbulence structure such as the time-averaged velocity profiles and fluctuation velocity profiles were precisely measured by using the constant temperature hot-wire anemometer. For the liquid film flow, the time-averaged film thickness, base-film thickness and interfacial wave height were obtained by using the point electrode resistivity probe. Direct observations for the interfacial waves on liquid film flow such as disturbance waves and ripple were also carried out by using the high-performance camera system to make clear the interfacial wave structure.

Commentary by Dr. Valentin Fuster
2005;():281-285. doi:10.1115/FEDSM2005-77407.

An experimental investigation of the effects of turbulence on primary breakup of round liquid jets subjected to gaseous crossflow is described. The paper investigates the effects of partial degrees of turbulence development in the liquid. Measurements of the properties of primary breakup were obtained using double-pulsed shadowgraphy in a subsonic wind tunnel having a test section of 0.3 m × 0.3 m × 0.6 m. Measurements included primary breakup regimes, conditions required for the onset of breakup, ligament properties along the liquid surface, drop size and velocity distributions after breakup along the liquid surface, conditions required for breakup of the liquid jet as a whole, and liquid jet trajectories.

Topics: Turbulence , Jets
Commentary by Dr. Valentin Fuster
2005;():287-292. doi:10.1115/FEDSM2005-77459.

The Couette flow of two immiscible liquids is examined using flow visualization techniques. The flow dynamics are studied as a function of several independent parameters including gravity. The two fluids are initially separated by a sheet of aluminum sufficient in length to ensure that fully developed flow conditions are achieved for both fluids before they come in contact with each other. The experiments are performed for various flowrates of Canola Oil and Polyethylene Glycol (PEG) corresponding Reynolds numbers for Oil and PEG of 0 to 20 and 0.01 to 0.2 respectively. Photographic images of the flow field are recorded and analyzed with the aid of image analysis software to illustrate interfacial dynamics of the flow. A qualitative and quantitative analysis of the flow instability is performed for various inclinations of the test apparatus, including the extreme cases of upward vertical and downward vertical with the horizontal being the baseline test case. Neutral stability curves are identified for the range of variables studied in the experiments. The long wave instability is observed to be very periodic. At the onset of instability, the flow structure is three-dimensional and exhibits wave growth in the flow direction. The wave growth ultimately results in droplet pinch off from the crest of a folded wave. At a constant relative velocity, the wave length is at a minimum when the flow is oriented in the upward vertical direction, opposing gravity. For a given PEG flowrate, the critical Oil flowrate for the onset of interfacial instability decreases as the angle increases. These results indicate gravity enhanced Kelvin-Helmholtz interfacial instability even for low Reynolds numbers. Through a course of systematic variation of flow angles we have been able to separate the effects of inertia, gravity (buoyancy) and viscous shear forces on the wavelength of instability.

Commentary by Dr. Valentin Fuster
2005;():293-297. doi:10.1115/FEDSM2005-77469.

A computational study of the deformation and surface wave properties of nonturbulent round liquid jets in gaseous crossflow is described. The objective of the study was to consider effects of liquid viscosity, liquid/gas density ratio, and crossflow Weber number that are representative of practical sprays. Three-dimensional computations of the deformation of round liquid jets in gaseous crossflow were carried out using FLUENT software utilizing its Volume of Fluid (VOF) module. The computations were evaluated satisfactorily based on earlier measurements of the properties of nonturbulent round liquid jets in crossflow (liquid jet deformation and surface waves) and revealed three-dimensional properties of the surface waves that could not be observed by previous measurements that were taken from the side of the jet.

Commentary by Dr. Valentin Fuster
2005;():299-303. doi:10.1115/FEDSM2005-77493.

Compressive surface-normal velocity gradient at a free surface leads to high mass transfer across a free surface. Our research aims to directly measure this velocity gradient at the free surface by proposing an advanced Particle Image Velocimetry (PIV) technique and simultaneously evaluate its applicability. This technique, PIV/IG (Interface Gradiometry), was proposed by Nguyen et al. (2004) to directly measure wall velocity gradient with high S.N.R. Herein, we adapt this technique to measure the compressive surface-normal velocity gradient at the free surface of open channel flow with minimal fluctuation of water surface. We validate this technique in a two-component PIV configuration by synthetic PIV images corresponding to uniform compression, linearly-varying compression, and a velocity field based on DNS data of open channel flow at friction Reynolds number Reτ = 240 and zero Froude number. The results clearly show that this technique works much better than the velocity differentiation method. The effect of template size on the measured value is evaluated.

Commentary by Dr. Valentin Fuster

Forum on Advances in Fluids Engineering Education

2005;():305-310. doi:10.1115/FEDSM2005-77191.

Propellers represent an interesting application of the principles of aerodynamics. The basic physics of propeller operation can be modeled as a rotating wing section using classical blade element analysis procedure, which can also include flows induced by the propeller motion itself. Performance testing of small-scale powered propellers in modest size educational wind tunnels could yield important verification of these analysis tools, and also provide valuable experimental insight into important aspects of propeller design for the engineering laboratory. To provide useful data, measurements of propeller performance must include not only rotation speed and thrust, but also torque. These variables need to be investigated as a function of the imposed wind tunnel airspeed, which represents the forward speed of a powered propeller in flight. Rotation speed is easily measured using a variety of simple optical (including stroboscopic) techniques and thrust simply corresponds to the axial force measurement obtained directly from the typical “sting” balance used with educational wind tunnels. However, commercial devices for practical torque measurement can be quite expensive and are also typically of much higher torque range than that achieved by small-scale propellers designed for model airplane use, which limits their usefulness in the educational engineering wind tunnel laboratory. This paper presents a simple and inexpensive strain gage based device designed for measurement of low level torque developed by small-scale powered propellers. The operating principles of the torque measurement device are described, along with static calibration test results and experimental measurements of the performance characteristics of a small-scale electric motor driven powered propeller using our educational wind tunnel test facility. The torque sensor can be combined with rapid prototyping propeller design to allow investigation of a wide variety of propeller design features. Additional planned improvements and other wind tunnel applications for the torque measurement device are also discussed in the paper.

Commentary by Dr. Valentin Fuster
2005;():311-317. doi:10.1115/FEDSM2005-77216.

Compressible flow is an important subject in aerospace and mechanical engineering disciplines. This paper describes the development of a web-base interactive compressible flow solver using Java programming language. The main objective of the solver is to provide students with a software tool than can be used in the compressible flow course offered in the Department of Mechanical Engineering at Lamar University. The solver has a graphical user interface (GUI) for ease of use and interactivity. The solver was developed with the intention of free distribution to the educational community and other interested users.

Commentary by Dr. Valentin Fuster
2005;():319-321. doi:10.1115/FEDSM2005-77336.

This paper discusses an undergraduate fluid mechanics laboratory session. The lab allows the students to observe various sediment transport phenomena in a hands-on manner. The experiments are performed in a glass-walled, tilting sediment flume. The following sediment transport phenomena are created and observed by the students — bed load, suspended load, bed forms (ripples, dunes, antidunes...), surface waves over various bed forms and local scour at flow obstructions including bridge piers and abutments. Students are able to observe local scour using PVC pipes for bridge piers and dimension lumber for abutment scour. Since the flume is 12.2-m long, a large group of students can spread out along both sides of the flume to observe bed forms and to perform local scour tests.

Commentary by Dr. Valentin Fuster

Forum on Transport Phenomena in Mixing

2005;():323-328. doi:10.1115/FEDSM2005-77051.

The mass flow passing through a plane normal to the mainstream direction of a free jet is the sum of the mass flow of the jet and that of the fluid entrained from surrounding ambient into which it issues. Manipulation of the instantaneous Navier-Stokes equations and the continuity equation yields an integro-differential equation for the instantaneous mass flow in the flow field. This equation is reduced to a form that suggests that jet entrainment may be viewed as a one-dimensional unsteady diffusion process with an integral source term arising from the gradient of forces in the axial direction of flow which are dependent, in general, on z and t. The small difference in the integrals of the net axial inertial force acting on the fluid in the volume defined by the limits of integration is balanced by an axial force arising from the viscous normal stress that is associated with axial rate of mass entrainment. Furthermore, it suggests that the kinematic viscosity of the fluid is the appropriate diffusion parameter. This formulation is used to assess the nature of the entrainment process in steady three-dimensional jets and to propose means for managing that process.

Topics: Jets
Commentary by Dr. Valentin Fuster
2005;():329-335. doi:10.1115/FEDSM2005-77079.

Modern power plants discharge approximately 1.5 to 2kWhr of waste heat for every kWhr of electrical energy produced. Modern power plants discharge approximately 1.5 to 2kWhr of waste heat for every kWhr of electrical energy produced. Usually this heat is discharged to an adjacent water body which increases the water temperature near the outfall. In order to assess the ecological consequences of waste heat discharge one must first know the physical changes (temperature, velocity, salinity) induced by these discharges. It is with this later aspect, prediction of physical properties, that the current work is primarily concerned. Existing theoretical work on axisymmetric buoyant jets is confined to integral techniques developed by Morton in the early 1950’s. From these techniques only centerline velocities and temperatures can be calculated. Experimental data for this type of flow are essentially confined to centerline temperature measurements except for pure jet or plume data which constitute the extremes for a buoyant jet. The present work addresses the problem of developing a theoretical model for an axisymmetric laminar buoyant jet. The governing equations for an axisymmetric buoyant jet in rectangular co-ordinates are transformed into an orthogonal curvilinear co-ordinate system which moves along the length of the jet axis. The complete partial differential equations governing steady, incompressible laminar flow are solved in the new curvilinear co-ordinates using finite-difference techniques. This method is applicable to a much wider range of jet flows issued at arbitrary angles into quiescent or flowing ambience. This method is also applied to the case for multiple jets spaced by a finite distance apart. Results for the momentum jet, axial and radial distribution of velocity and temperature, show good agreement with published data.

Topics: Jets
Commentary by Dr. Valentin Fuster
2005;():337-345. doi:10.1115/FEDSM2005-77221.

Present work studies the mixing in turbulent flow behind a backward-facing step, which is a classical example of mixing of jet in cross flow. In this study the cross flow was an unconfined cross flow flowing through a duct, which has geometry of backward-facing step with varying cross section. Three jet geometries, a slot jet, a single circular jet and three circular jets with different injection angles were studied. Spatial unmixedness based on helium mass concentration was used as an indicator for determining the degree of mixing. Velocity ratio, defined as ratio of jet velocity to the cross flow velocity, was varied during the analysis. Three different values of velocity ratio were considered during the study. Computational fluid dynamics analysis was performed using commercial CFD software CFX 5.6. Results were extracted in the form of helium mass concentration fields, concentration profiles, concentration contour plots and velocity vector plots. From the results obtained it can be concluded that, angle of jet inclination is the important controlling parameter in mixing in flow behind a backward-facing step. Single jet geometry gives better mixing for the given conditions.

Commentary by Dr. Valentin Fuster
2005;():347-352. doi:10.1115/FEDSM2005-77401.

Characteristics of turbulent flow and mass transfer around a rotating circular cylinder are investigated by Direct Numerical Simulation. Mass-transfer results are presented at a high Schmidt number (Sc = 1670). The concentration field is computed for three different cases of low Reynolds number, Re*R = 161, 348 and 623, based on the cylinder radius and friction velocity. Results confirm that the thickness of Nernst diffusion layer is very small compared with that of viscous sub-layer in the case of high Sc mass transfer. A strong correlation of the concentration field with streamwise and vertical velocity components is noticed. However, that is not the case with the spanwise velocity component. Visualization of instantaneous concentration reveals that the length scale of concentration fluctuation typically decreases as Reynolds number increases. The correlation between Sherwood number and Reynolds number is consistent with other experiments currently available.

Commentary by Dr. Valentin Fuster
2005;():353-360. doi:10.1115/FEDSM2005-77454.

A numerical approach to optimizing heat removal from a standing wave thermoacoustic engine by generating steady circulation at the outlet is described. The open-ended standing wave engine makes use of a hollow cone placed opening outwards and flush with the open end of the engine. Parallel paths within and without the cone create asymmetric minor losses. These are designed to induce circulation and heat removal. If successful, this will eliminate the need of a cold heat exchanger. The numerical models make use of the commercial RANS numerical solver FLUENT. Orders of magnitude increase in heat flux out of the engine’s open tube as compared to an engine without a cone are demonstrated. Three cones are evaluated at a range of flow parameters. Performance is found to be a strong function of the displacement amplitude compared to the engine diameter. In addition, these results are compared to an experiment that provides velocity data.

Commentary by Dr. Valentin Fuster

14th Forum on Industrial, Environmental, and Petrochemical Applications of Fluid Mechanics

2005;():361-367. doi:10.1115/FEDSM2005-77383.

Analyses of water jet based on two-fluid model of two-phase dispersed flow have been carried out for single water jet and cross water jet in relation to the water jet technology in civil engineering. Mass and momentum conservation equations for liquid phase (droplet) gas phase (air) were formulated separately (two-fluid model formulation). Physical modeling of diffusion of droplets, drag coefficient of droplet in dispersed flow, shear stress at jet interface, etc has been carried out in detail and constitutive equations for these physical phenomena have been developed. Based on the two-fluid model basic equations and constitutive equations, one-dimensional analysis has been carried out considering simplified mode and hydrodynamic structure of water jet was well predicted. In the practical application of present analyses, some preliminary analyses on cross jet where two water jets collide with certain collision angles have been carried out and the predicted results reasonably explain the experimental results.

Topics: Soil , Water
Commentary by Dr. Valentin Fuster
2005;():369-372. doi:10.1115/FEDSM2005-77386.

The aim of present study is to establish the numerical model for the solidification or melting of water saturated soil and to clarify the effect of thermal and physical parameters on the artificial soil freezing by comparing between the numerical and experimental results. First, the numerical model has been modified to adapt for freezing of soil. By comparing between obtained numerical solutions and experimental data, the validity of the model has been checked and certified. Next, the effect of physical property of soil, initial and boundary conditions of soil and freezing pipes, the velocity of groundwater, and the arrangement of freezing pipes on soil freezing have been examined. As the results, it was found that the water content of soil and ground water affect the volume of solid, besides the groundwater also especially changes the profile of solid/liquid interface. The rate of the interface growth would gradually stop provided that the flow speed exceeds certain limits. The knowledge obtained from our study will be useful to predict solid volume, decrease in thermal energy consumption and minimize the influence to ambience on artificial ground freezing precisely.

Topics: Freezing
Commentary by Dr. Valentin Fuster
2005;():373-377. doi:10.1115/FEDSM2005-77410.

A new nozzle using a composite flow network has been developed for plain air film stabilization without film break. The design consists of a slit duct with layers of diamond-shaped cylinder bundles. Tests were conducted using air at average velocity of 8m/s. The new nozzle had a stabilized length in its flow direction stretching 2 times, compared to a conventional slit design. The development of new nozzle was based on fluid dynamic theory. A new type of intersecting flow networks, to be referred to as composite flow network, was constructed consisting of ambivalent flows superposed with con-current thin-film flows in diamond-shaped cylinder bundles. Our previous experiments have revealed the occurrence of flip-flop flow with self-sustained flow oscillations and the generation of longitudinal vortices in the diamond-shaped cylinder bundle. It is conceivable to extend longitudinal vortex flow in composite flow networks through interacting with flip-flop flow exiting from a pre-fixed multiple-intersecting flow network in order to strengthen and thus stretch its efflux.

Commentary by Dr. Valentin Fuster
2005;():379-387. doi:10.1115/FEDSM2005-77476.

A jet pump may be constructed from a pair of concentric tubes in which the center tube is shaped as a converging-diverging nozzle. Primary fluid is allowed to accelerate through the nozzle, thus creating a low-pressure region at the nozzle exit. Secondary fluid flowing in the peripheral region is drawn into the low-pressure region and is thus accelerated. In this study, the jet pump is employed as part of a space thermal-management system based on a cycle known as the Solar Integrated Thermal Management and Power (SITMAP). The latter is a combined vapor compression cycle and a Rankine cycle with the compression device being a jet pump instead of the regular compressor. The jet pump has several advantages for space applications as it involves no moving parts, a feature that results in decreasing the weight and vibration level while increasing the system reliability. The working fluid is cryogenic nitrogen, which is readily present onboard the spacecraft. This study presents a detailed component analysis of the jet pump allowing for two-phase supersonic flow. The model also accounts for Fabri choking either at the inlet or at an aerodynamic throat in the mixing chamber. The model also accounts for flow choking at the exit of the mixing chamber. The different choking situations limit the ability of the jet pump to entrain more secondary flow once the flow is choked at any location. In this study the various choking regimes will be identified and the entrainment ratios corresponding to the different choking scenarios will be calculated.

Commentary by Dr. Valentin Fuster

Forum on Fluid Measurements and Instrumentation

2005;():389-396. doi:10.1115/FEDSM2005-77121.

Within the Reactor Pressure Vessel (RPV) plant life time project, Electricité de France (EDF) must carry out studies showing the RPV integrity during severe loading conditions as the Small Break Loss Of Coolant Accidents (SBLOCA) scenarios. A recent thermal-hydraulic SBCLOCA scenario study, concerning a certain RPV geometry, showed two-phase-flow conditions appears in the cold leg. The single phase flow CFD tools, usually used in these cases, are unsuitable to carry out such studies. That is the reason why EDF is developing a new CFD code (called Neptune_Code) implementing two-phase flow models. In order to improve the physical phenomena knowledge and the existent experimental data for Neptune_Code development, EDF carried out an experimental study concerning the stratified flow (salt water) and free surface (water air) behavior during a representative ECC scenario with partial uncovered cold leg. This paper focuses mainly on the experimental method and the measurements obtained in three tests showing the effects of the uncovered cold leg level on the stratified flow and the free surface behavior.

Commentary by Dr. Valentin Fuster
2005;():397-402. doi:10.1115/FEDSM2005-77254.

A Global Laser Range Profilometry (GLRP) System has been developed at the Naval Surface Warfare Center, Carderock Division (NSWCCD) for the measurement of three dimensional wave surfaces. A laser diode array consisting of 100 diodes operating at 650nm with an energy output of 3.5 mW was used to illuminate the wave surface seeded with fluorescent dye at various points in the Miniature Water Basin (MWB) facility at NSWCCD. A CCD camera located above the water surface recorded successive images of the laser array at 30 frames per second. Image processing techniques were used to locate the centroid of the laser array diode locations in the collected images. Calibration images were taken at various calm water heights in order to convert the image centroid locations to real coordinates. Two- and three-dimensional results are presented, along with error analysis of the GLRP system and comparison to flow visualization. A discussion of possible applications and planned future work is also presented in this work.

Topics: Lasers , Waves
Commentary by Dr. Valentin Fuster
2005;():403-408. doi:10.1115/FEDSM2005-77284.

This experimental study, applied to a three different sweep axial fan (backward, radial and forward), aims at determining the 3D structure of the rotor wake from unsteady velocity measurements. The hot-film anemometry is used to measure the 3D unsteady velocity components in nearfield, downstream the fan. The data analysis leading to averaged and turbulent velocities, the components of the Reynolds’ stress tensor and the turbulent kinetic energy is presented, in order to illustrate the influence of the sweep. A spectral analysis is also performed.

Topics: Wakes , Rotors
Commentary by Dr. Valentin Fuster
2005;():409-414. doi:10.1115/FEDSM2005-77349.

The orifice plate flow meter is the most common form of differential pressure flow meter used in industry. The standard discharge coefficient, which is defined by both British Standard and ISO 5167, is only valid if the flow approaching the meter is perfectly settled and fully developed. However, in practical applications the flow approaching the orifice meter is often disturbed by pipe-fittings and consequently the measurements become inaccurate. Basically, the design of the orifice plate meters that are independent of the upstream disturbances is a main goal for orifice plate metering. This task can be achieved either by using a long straight settling length upstream and downstream of the orifice plate or using a flow conditioner upstream of an orifice plate. In addition, the standard orifice plate is vulnerable when metering dirty flow due to the accumulation of dirt in front of the orifice plate which can alter the accuracy of metering as well. In this paper the effect of the swirler flow conditioner for both standard and non-standard flow conditions has been investigated in an experimental rig and validation of the results has been justified with the appropriate CFD domains. In these investigations the effect of different designs of swirler flow conditioners have been examined in asymmetric and swirling flow profiles. The results so far show the cone swirler flow conditioner has a desirable effect for both asymmetric and swirling flow disturbances. They also show the error of metering for non-standard velocity profiles with the swirler flow conditioner is typically 1.5% compared to around 4% for a standard orifice plate. Moreover using a swirler conditioner tends to keep particles in suspension and thus prevents the accumulation of dirt particles in front of orifice plate. All experimental and numerical results here are presented for different velocity profiles both swirling and asymmetric profiles, mass flow rates and for β = 0.5.

Commentary by Dr. Valentin Fuster
2005;():415-423. doi:10.1115/FEDSM2005-77359.

Gas-liquid, gas-solid, liquid-solid, and gas-liquid-solid multiphase flows are difficult to visualize, characterize, and quantify because the systems are typically opaque. Invasive or noninvasive measurement methods are typically used for determining internal flow and transport characteristics of these complex flows. The difficulty with invasive methods is that they can alter the internal flow of a multiphase system causing interference with realistic process measurements. X-ray imaging provides one family of noninvasive measurement techniques used extensively for product testing and evaluation of static objects with complex structures. These techniques have been extended to visualize dynamic systems, such as those which characterize multiphase flows. This paper will describe a new X-ray flow visualization facility for large-scale multiphase flows. X-ray radiography and X-ray computed tomography of static and dynamic systems will be used to demonstrate system capabilities. Radiographic images will show bread dough rising, objects falling in a liquid, large bubbles rising in a 32 cm ID column of water, and operation of a 32 cm ID bubble column. X-ray computed tomography of a large static object will demonstrate visualization capabilities. X-ray computed tomography of a multiphase flow in a 32 cm bubble column will show local time-averaged gas holdup values for various operating conditions. Finally, challenges associated with X-ray stereographic imaging to capture time-resolved dynamic events will be outlined.

Commentary by Dr. Valentin Fuster
2005;():425-432. doi:10.1115/FEDSM2005-77377.

This is the second of two papers describing the traceability of nuclear feedwater flow measurements. The first considered the challenges and methodology for establishing the traceability of chordal ultrasonic flow meters. This paper considers the challenges of establishing the traceability in a measurement using a flow element of the modified venturi tube type. It specifically considers the assumptions and uncertainties associated with the extrapolation, for use in the field, of tube calibration factors measured in the laboratory. To quantify these uncertainties, the in-situ performance of four modified venturi tubes is compared with the performance of four 8-path chordal ultrasonic flowmeters. The data analyzed were collected in the feeds of four steam generators in a large pressurized water reactor plant, each feed containing one meter of each type. The meters were initially calibrated in this series arrangement in a NIST traceable calibration lab and then operated in the same arrangement in the field.

Commentary by Dr. Valentin Fuster
2005;():433-439. doi:10.1115/FEDSM2005-77394.

The objective of the work was to evaluate the efficiency of a hydraulic turbine by means of the flow measurement, for a given water head. The hydraulic turbine of 180 MW output has been in service for 20 years. The real value of efficiency was needed in order to proceed with minor/mayor modifications to improve it. In a case of a runner deterioration the pressure-time (the Gibson) method was chosen to proceed with a test for flow determination. However, to measure the pressure in the penstock no access from the external space of the penstock was found, so the special instrumentation had to be developed, which could be installed inside different sections of the penstock for determination of the pressure as required by the Gibson method. After the successful installation of the pressure transducers and a special hermetic capsule, from which a cable was laid through the manhole to the control room, the test was carried out at different loads applying the Gibson method. Simultaneously, the instrumentation for the Winter-Kennedy method was installed and calibrated during the test. In the paper all the turbine measured characteristics are given and discussed. It was concluded that the efficiency of the hydraulic turbine was still high and no modifications were necessary. Having instruments calibrated for the Winter-Kennedy method other curves can be obtained at different heads.

Commentary by Dr. Valentin Fuster
2005;():441-445. doi:10.1115/FEDSM2005-77402.

This paper presents new experimental data for orifice meter expansion factor tests flowing natural gas through a 75 mm (3-inch) diameter orifice meter tube for values of the orifice diameter ratio (β) between 0.60 and 0.75. The tests were performed in the Low Pressure Loop (LPL) of the Metering Research Facility (MRF) at Southwest Research Institute (SwRI). The new data complement previously reported results for 100 mm (4-inch) and 150 mm (6-inch) meter tubes for β values between 0.20 and 0.60.

Topics: Pressure , Natural gas
Commentary by Dr. Valentin Fuster
2005;():447-451. doi:10.1115/FEDSM2005-77403.

Multi-path gas ultrasonic flow meters are used to measure the flow rate of natural gas in custody-transfer metering applications. Steady-flow tests were performed in the high-pressure loop (HPL) of the Southwest Research Institute (SwRI) Metering Research Facility (MRF) flowing natural gas through two 300 mm (12-inch) diameter multi-path ultrasonic flow meters with different ultrasonic path configurations. Tests were performed with both small and large temperature differences between the flowing gas temperature and the outdoor ambient temperature. This paper presents the results of the large temperature difference tests with and without an upstream flow conditioner for one multi-path ultrasonic meter in the low-flow range of 0.15 m/s (0.5 ft/s) to 0.30 m/s (1 ft/s). Test conditions were selected to complement a computational fluid dynamics (CFD) study performed by Morrison and Brar [2004,2005] at Texas A&M University. The experimental results confirm that the gas flow in the ultrasonic meter was thermally stratified (as predicted by Morrison and Brar [2004]) and show the effects of thermal stratification on path velocities, meter diagnostic path velocity ratios, and on meter accuracy. The results show that the flow conditioner was relatively ineffective in smoothing the axial velocity profile distortion caused by thermal stratification in this low velocity range.

Topics: Flowmeters
Commentary by Dr. Valentin Fuster
2005;():453-457. doi:10.1115/FEDSM2005-77423.

High-speed in-line digital holographic cinematography was used to investigate the diffusion of droplets in locally isotropic turbulence. Droplets of diesel fuel (0.3–0.9mm diameter, specific gravity of 0.85) were injected into a 37×37×37mm3 sample volume located in the center of a 160-liter tank. The turbulence was generated by 4 spinning grids, located symmetrically in the corners of the tank, and was characterized prior to the experiments. The sample volume was back illuminated with two perpendicular collimated beams of coherent laser light and time series of in-line holograms were recorded with two high-speed digital cameras at 500 frames/sec. Numerical reconstruction generated a time series of high-resolution images of the droplets throughout the sample volume. We developed an algorithm for automatically detecting the droplet trajectories from each view, for matching the two views to obtain the three-dimensional tracks, and for calculating the time history of velocity. We also measured the mean fluid motion using 2-D PIV. The data enabled us to calculate the Lagrangian velocity autocorrelation function.

Commentary by Dr. Valentin Fuster
2005;():459-465. doi:10.1115/FEDSM2005-77437.

The Peak-locking effect causes mean bias in most of the existing correlation based algorithms for PIV data analysis. This phenomenon is inherent to the Sub-pixel Curve Fitting (SPCF) through discrete correlation values, which is used to obtain the sub-pixel part of the displacement. A new technique for obtaining sub-pixel accuracy, the Correlation Mapping Method (CMM), was proposed by Chen & Katz [1, 2]. This new method works effectively and the peak-locking disappears in all the previous test cases, including applying to both synthetic and experimental images. The random errors are also significantly reduced. In this paper, an optimization of the algorithm is reported. Using sub-pixel interpolation, the cross-correlation function between image 1 and image 2 is expressed as a polynomial function with unknown displacement, in which the coefficients are determined by the autocorrelation function of the image 1. This virtual correlation function can be matched with the exact correlation value at every point in the vicinity of the discrete correlation peak (a 5×5 pixels area is chosen in the present study). A least square method is used to find the optimal displacement components that minimize the difference between the real and virtual correlation values. The performances of this method at the presence of background noise and out-of-plane motion are investigated by using synthetic images, as well as the influence of under-resolved particle images, and compared with the result of the SPCF method. The advantage of the CMM over SPCF is demonstrated in these studies.

Commentary by Dr. Valentin Fuster
2005;():467-474. doi:10.1115/FEDSM2005-77470.

The effect of the inlet curvature R on the discharge coefficients of toroidal throat critical flow Venturi nozzles is discussed based on calibration results of high-precision nozzles HPNs with R = 1.0, 1.2, 1.5, 1.8, 2.0, and 2.5D, where D is the throat diameter (9.6, 13.4, and 18.9 mm). The Reynolds numbers RD range from 1.2×105 to 1.2×106 , corresponding to absolute upstream pressures of 0.1 to 0.7 MPa. The calibrations were performed by a constant volume tank system developed for the primary standard in Japan.

Commentary by Dr. Valentin Fuster
2005;():475-482. doi:10.1115/FEDSM2005-77478.

The flow field inside gas pipeline meter run is numerically simulated to determine the affects of upstream piping and temperature differences between the meter run pipe and the gas upon the flow. At bulk averaged velocities below 0.6 m/s (2 ft/s) significant changes in the velocity field are present which may alter the response of any flow meter mounted in the meter run. Examples for a bulk average velocity of 0.15 m/s (1/2 ft/s) and temperature differences with magnitudes of 27.7°C (50°F) are presented.

Commentary by Dr. Valentin Fuster
2005;():483-492. doi:10.1115/FEDSM2005-77484.

This paper presents the first implementation of a novel class of dynamic time-resolved direct skin friction measurements sensor based on active ionic polymer transducers. These ionic polymer sensors have the advantage that they contain no moving parts, perform a direct measurement of shear, and can be mounted directly to the surface of an existing vessel with no modification. During the present effort we characterize the accuracy of the sensors and validate their dynamic measurement response. Using an oscillating Stokes layer calibration procedure we demonstrate measurement accuracy in fluctuating shear on the order of 4.92% over a range of stresses of +/- 3 Pa and signal-to-noise-ratio on the order of 60 dB. The frequency response of the sensor is over 10 kHz however due to experimental limitations we were not able to calibrate for frequencies higher than 140 Hz. These sensors have been shown to be insensitive to vibration or pressure. Also, an automatic change of impedance compensation approach is proposed that allows in-situ recalibration of the sensors and accounts for environmental effects such as changes of temperature on the sensors performance. The results demonstrate the potential for using ionic polymer sensors to perform accurate, high frequency measurements of shear in turbulent boundary layers.

Commentary by Dr. Valentin Fuster
2005;():493-498. doi:10.1115/FEDSM2005-77486.

Applying similarity analysis to the RANS equations of motion for a pressure gradient turbulent boundary layer, Castillo and George [1] obtained the scalings for the mean deficit velocity and the Reynolds stresses. Following this analysis, Castillo and George studied favorable pressure gradient (FPG) turbulent boundary layers. They were able to obtain a single curve for FPG flows when scaling the mean deficit velocity profiles. In this study, FPG turbulent boundary layers are analyzed as well as relaminarized boundary layers subjected to an even stronger FPG. It is found that the mean deficit velocity profiles diminish when scaled using the Castillo and George [1] scaling, U∞ , and the Zagarola and Smits [2] scaling, U∞ δ*/δ. In addition, Reynolds stress data has been analyzed and it is found that the relaminarized boundary layer data decreases drastically in all components of the Reynolds stresses. Furthermore, it will be shown that the shape of the profile for the wall-normal and Reynolds shear stress components change drastically given the relaminarized state. Therefore, the mean velocity deficit profiles as well as Reynolds stresses are found to be necessary in order to understand not only FPG flows, but also relaminarized boundary layers.

Commentary by Dr. Valentin Fuster
2005;():499-504. doi:10.1115/FEDSM2005-77488.

A real time multi-functional ultrasonic sensor system is proposed to provide automated drilling fluid monitoring that can improve the capability and development of slimhole and microhole drilling. This type of reliable, accurate, and affordable drilling fluid monitoring will reduce the overall costs in exploration and production. It will also allow more effective drilling process automation while providing rig personnel a safer and more efficient work environment. Accurate and timely measurements of drilling fluid properties such as flow rate, density, viscosity, and solid loading are key components for characterizing rate of drill penetration, providing early warning of lost circulation, and for use in real-time well control. Continuous drilling fluid monitoring enhances drilling economics by reducing the risk of costly drilling downtime, increasing production performance, and improving well control. Investigations conducted to characterize physical properties of drilling mud indicate that ultrasound can be used to provide real-time, in-situ process monitoring and control. Three types of ultrasonic measurements were evaluated which include analysis of in wall, through wall and direct contact signals. In wall measurements provide acoustic impedance (the slurry density and speed of sound product). Through wall and direct contact measurements provide speed of sound and attenuation. This information is combined to determine physical properties such as slurry density, solids concentration and can be used to detect particle size changes and the presence of low levels of gas. The measurements showed that for the frequency range investigated in-wall measurements were obtained over the slurry density range from 1500 to 2200 kg/m3 (10 to 17 pounds solids per gallon of drilling fluid). Other measurements were obtained at densities in the 1500 to 1800 kg/m3 range. These promising measurement results show that ultrasound can be used for real-time in-situ characterization of the drilling process by monitoring drilling mud characteristics.

Topics: Drilling
Commentary by Dr. Valentin Fuster

Cavitation and Multiphase Flow Forum

2005;():505-512. doi:10.1115/FEDSM2005-77113.

Since cavitation is often unavoidable, it is crucial to understand how fluid-handling machinery operates under cavitating conditions. The purpose of this research is to investigate two-phase flow structure in the wake of a hydrofoil undergoing unsteady partial cavitation using an integrated experimental and numerical approach. Experiments were conducted systematically in the high-speed water tunnel at the St. Anthony Falls Laboratory to capture the characteristics of the bubbly wake induced by unsteady partial cavitation. The velocity and bubble information were obtained using a TSI color burst LDA/PDA system with a beam splitter and IFA processor. A nominal sampling period of 30 seconds per coordinate point was chosen for the collection of all data sets of LDV measurements. Since the LDV data rate changes with position, the collection period was also controlled by the upper limit of samples, which were 3,000 samples per point. In order to get enough samples, a nominal sampling period of 180 seconds per coordinate point was selected for PDA measurements and 10,000 samples were used as the upper limit. The actual number of samples for each PDA measurement location was between 5,000 and 10,000 with an average data rate of 40 Hz. The averaged bubble size measured in the wake is obtained to be 240 ∼ 300 microns and the estimated maximum void fraction in the wake being in the order of one percent. A virtual single-phase natural cavitation model with the effect of incondensable gas was developed and implemented to better understand the cavitating wake physics. A number of parameters, such as time-averaged velocity distributions both on suction side of the hydrofoil and in the wake, spectral characteristics of unsteady lift oscillations etc., were computed and found in very good agreement with experimental data (Qin 2004), indicating this model can capture the main physics of unsteady cavitating flow. It was also found that the peak of time-averaged incondensable gas in the wake is in the order of one percent, indicating that the majority of vapor bubbles were condensed to water in the wake and hence the bubbles in the wake of unsteady cavitation are mainly gas bubbles.

Topics: Wakes , Hydrofoil
Commentary by Dr. Valentin Fuster
2005;():513-517. doi:10.1115/FEDSM2005-77114.

Cloud cavitation is the rapid formation and shedding of vaporous clouds from a cavitating hydrofoil. This type of cavitation occurs under certain conditions that are characterized by the cavitation number [σ] and the angle of attack [α]. Associated with cloud cavitation are large, abrupt changes in surface pressure caused by the shedding of the attached cavity. Our experimental data display trends that are contained in the linearized flat plate theories of Acosta and Tulin. Near values of σ/2α equal to 4, a singularity exists in the flat plate theory. Experimental results and numerical simulations indicate that in this region a transition between competing mechanisms of cavity shedding occurs. A new finding is that water quality appears to have a significant effect on cavitation behavior. It is well known that nuclei content plays an important role in cavitation inception. However, a recent investigation made possible by high-speed video reveals that the cloud shedding is periodic and that, for each cycle, the cavitating surface becomes fully wetted. Thus, inception physics come into play for a fraction of each cycle. Experimental data shows that the fraction of time in each period that the hydrofoil is fully wetted varies with gas content. In addition, the spectral characteristics of lift and surface pressure measurements show a strong dependence on gas content. Numerical simulations made to incorporate gas content effects show surprisingly close agreement with experimental data. This is also factor that may be of importance in comparing the results from different experimental facilities since comparisons are often made without considering gas content as a factor.

Commentary by Dr. Valentin Fuster
2005;():519-524. doi:10.1115/FEDSM2005-77131.

The theory of cavitation in an ideal fluid is utilized to design hydrofoils that have a significant increase of lift to drag ratio for a regime of partially cavitating flows. Our recently reported experiments with natural cavitation have confirmed the existence of such an increase within a certain range of cavitation number and angle of attack for the specially designed hydrofoil designated as OK-2003. For applications of such a design to engineering, it would be necessary to keep the cavitation number within this favorable range and ventilation looks to be the most promising tool for control of cavitating flows. Therefore, comparative water tunnel tests have been carried out for both natural and ventilated cavitation of the OK-2003. The general similarity between the two kinds of partial cavitation for the developed low-drag hydrofoil is proven. When validating theory with the aid of water tunnel experiments, a general issue of how to make a comparison between natural cavitation and ventilated cavitation was encountered. This issue is the difficulty to determine the pressure within partial cavities. During natural cavitation the cavity pressure can deviate from vapor pressure due to the effects of dissolved gas and possibly other water quality effects. Direct pressure measurements within the partial cavity have proved to be unstable due to the unsteadiness of the cavity. The unsteadiness effect becomes more dominant as cavitation number is increased and the cavity becomes smaller. There is a point where the measured cavity pressure becomes unusable. In the case of ventilated cavitation, the interaction of the airflow with the surface of relatively thin cavities can be significant. Finally, it was experimentally determined that different dynamics of cavity pulsation are inherent to natural and ventilated cavitation.

Commentary by Dr. Valentin Fuster
2005;():525-530. doi:10.1115/FEDSM2005-77163.

This paper reports the results of measurements of the effective tensile strength Fc of water, in experiments involving a pulse of tension (‘negative pressure’) created by the reflection of a pressure pulse at a boundary, as a function of temperature. Using a modified ‘Bullet-Piston’ (B-P) pulse reflection apparatus, measurements presented herein show that degassed, deionised water is capable of sustaining tensions an order of magnitude greater than previously reported in B-P work. A theoretical explanation is developed indicating that the pressure records arising in B-P experiments are the result of cavitation due to a pulse of tension. Results are reported for measurements of Fc made over the temperature range 1°C ≤ T ≤ 95°C.

Commentary by Dr. Valentin Fuster
2005;():531-538. doi:10.1115/FEDSM2005-77200.

A detailed study of ventilated supercavitation in the reentrant jet regime is being carried out in the high-speed water tunnel at St. Anthony Falls Laboratory, as the hydrodynamics part of an interdisciplinary study on stability and control of high-speed cavity-running bodies. It is aimed at understanding the interaction between a ventilated supercavity and its turbulent bubbly wake, with the goal to provide the information needed for the development of control algorithms. Here Particle Image Velocimetry (PIV) measurements in high void fraction bubbly wakes created by the collapse of ventilated supercavities are reported. Bubble velocity fields are obtained, and shown to submit to the same high Reynolds number similarity scaling as the single-phase turbulent axisymmetric wake. A grayscale technique to measure local average void fraction is outlined. Initial results of a time-resolved PIV experiment (2000 Hz) are also presented.

Topics: Measurement , Wakes , Porosity
Commentary by Dr. Valentin Fuster
2005;():539-544. doi:10.1115/FEDSM2005-77271.

This paper surveys fluid dynamic-acoustic mechanisms that may explain low-frequent, broadband hull excitation experienced on board of ships and caused by propeller cavitation. Observations obtained from sea trials and model scale experiments are used to describe the hydrodynamics involved in each particular mechanism. The investigations are still ongoing and aim to identify causes of broadband inboard noise and vibration on passenger vessels in the frequency range of 20 to 70 Hz.

Commentary by Dr. Valentin Fuster
2005;():545-551. doi:10.1115/FEDSM2005-77352.

A study of visual and erosion effects of cavitation on simple single hydrofoil configurations in a cavitation tunnel was made. A two-dimensional hydrofoil with circular leading edge was used for the experiments. In addition, the hydrofoil geometry was modified to obtain some three-dimensional cavitation effects. A thin copper foil, applied to the surface of the hydrofoil, was used as an erosion sensor. The cavitation phenomenon on hydrofoils at different flow conditions (system pressure, water gas content, flow velocity) was observed. Images of vapour cavities from above and from a side view were taken. Plausible results that showed a significant relationship between cavitation erosion and the visual effects of cavitation made it possible to use these information to develop a cavitation erosion model. The model is based on the physical description of different phenomena, which are involved in the process of pit formation — pressure wave emission and its attenuation, micro-jet formation, the jet impact to the solid surface and pit creation. The model is capable to predict the influence of significant parameters as flow velocity and gas content of water. The model that was developed on the basis of measurements of cavitation on a single hydrofoil was later tested on an actual hydraulic machine in the form of a radial pump. The predicted magnitude and distribution of cavitation damage relates well to the experimentally measured one.

Commentary by Dr. Valentin Fuster
2005;():553-560. doi:10.1115/FEDSM2005-77368.

The experimental data which will be presented in this paper are the results of the comparison between different methods for evaluating damaged surfaces by cavitation erosion. The different methods are partly working in the initial stage of cavitation erosion and partly at developed cavitation erosion, where mass loss occurs. The used test rig consists basically of a rotating disc with a diameter of 500 mm on which four holes are located. Each hole generates a cavitation zone while the disc is rotating. The test objects are material specimens made of copper. Copper was used as test material in respect to reasonable durations for the tests. The specimen can be implemented in the casing of the test rig directly across the rotating disc on the diameter where the holes are located. This rotating disc test rig generates a very aggressive type of cavitation, so that mass loss, of course depending on the tested material, will appear after relatively short durations. Also the initial stage of cavitation erosion can be observed. The used test rig is very interesting regarding the possibility to apply different measuring techniques to characterize the erosive aggressiveness of cavitation. These techniques are at first the so-called Pitcount-Method, which allows investigations of cavitation erosion in the initial stage. The second one is an acoustic method, which is based on a structure-borne noise sensor and a specially developed signal processing system. The third method is the measuring of mass loss of the material specimen after several time steps. With the help of a CCD-camera and special digital image processing software, images of different cavitation conditions were recorded. The information obtained from these images should serve as support for the evaluation of the other used methods. After the evaluation with the above mentioned methods, the specimens were evaluated with a special device which works with the help of a white light interferometer. With this evaluation method three-dimensional information can be obtained in respect to the actually eroded volume of the specimens. With this information the lost mass of the specimens could be calculated directly. Especially the comparison of the results obtained from the Pitcount-Method, which is a two-dimensional evaluation method, and the three-dimensional results of the white light interferometer is an important point of the work within this paper.

Topics: Cavitation
Commentary by Dr. Valentin Fuster
2005;():561-566. doi:10.1115/FEDSM2005-77371.

This work intended to evaluate the instantaneous vapour fraction in the turbo-pump inducer of a liquid propellant rocket engine. Experimentations held on an experimental pump test facility and cavitation was attained by reducing the inlet pressure in the machine while maintaining constant the inducer rotational speed. Measurements of vapour fraction through the rotating inducer were achieved by means of an x-ray-based system. The system exerted an industrial x-ray generator and 10 collimated scintillation detectors. Detectors were functioning in current mode thus permitting an acquisition at 5 kHz for each detector. A reference x-ray detector situated between the x-ray generator and the machine permitted the treatment of x-ray beam energy fluctuations related to industrial generators [1, 2, 3]. Acquisitions were performed in three axial positions on the inducer. For each measurement position, three cavitation sequences with different flow rate conditions (Q/Qn = 0.9, 1, 1.1, where Qn is the nominal flow rate) were accomplished. Each cycle is performed by decreasing gradually the pressure while maintaining an imposed rotational speed of 4000 rpm. Each test is constituted of 10 pressure points varying from 2.40 to 0.48 bars representing a complete cavitation sequence. X-ray acquisition was performed for each pressure point and it was carried out for 10 seconds thus corresponding to 667 tours of the inducer. Vapour fraction was determined instantaneously thus showing the applicability and the precision of the method in such measurements despite of the geometry and rotation speed constraints. Consequently a quantitative and qualitative evaluation of the vapour fraction is presented. Results show that the vapour distribution is well related to cavitation development on the blades of the inducer for steady cavitation condition.

Topics: X-rays
Commentary by Dr. Valentin Fuster
2005;():567-573. doi:10.1115/FEDSM2005-77387.

It is well known that the suction performance of turbopumps in cryogenic fluids is much better than that in cold water because of thermodynamic effect of cavitation. In the present study, an analytical method to simulate partially cavitating flow with the thermodynamic effect in a cascade is proposed; heat transfer between the cavity and the ambient fluid is modeled by one-dimensional unsteady heat conduction model under the slender body approximation and is coupled with a flow analysis using singularity methods. In this report, the steady analysis is performed and the results are compared with those of experiments to validate the model of the present analysis. This analysis can be easily extended into unsteady stability analysis for cavitation instabilities such as rotating cavitation and cavitation surge.

Topics: Cavitation
Commentary by Dr. Valentin Fuster
2005;():575-580. doi:10.1115/FEDSM2005-77397.

Propagation of pressure waves caused by a thermal shock in liquid metals containing gas bubbles is performed by a numerical simulation. The present study examined the influences of bubble radius and void fraction on the absorption of thermal expansion of liquid metals and attenuation of pressure waves. As the result of the calculation, since the large bubbles which have a lower natural frequency than the small bubbles cannot respond to the heat input, the peak pressure at the heated region increases with increasing bubble radius. Especially, when the bubble radii are around 500 μm, the pressure wave propagates through the mixture not with the sonic speed of the mixture but with that of liquid mercury. On the other hand, decreasing the void fraction makes behavior of bubbles nonlinear and a collapse of bubble produces a high pressure wave. However, the calculation shows that the method of introducing micro gas bubbles into liquid metals is effective to prevent cavitation erosion on the wall.

Commentary by Dr. Valentin Fuster
2005;():581-586. doi:10.1115/FEDSM2005-77405.

Hydrodynamic cavitation in micro systems is a fundamental engineering problem that is poorly understood. The lack of knowledge on cavitation relevant to MEMS devices is impeding the practical realization of novel high-velocity microfluidic machines. This paper divulges differences between cavitation occurring inside micro and conventional systems, and also indicates the limited applicability of conventional knowledge to predict and understand cavitating flows in micro-domains. A detailed discussion delineating the possible reasons of such a divergence is presented in this article. Additionally, selected results obtained from preliminary experiments on cavitation in micro-domains are presented.

Commentary by Dr. Valentin Fuster
2005;():587-590. doi:10.1115/FEDSM2005-77414.

CFD simulations were applied to cavitating flows around an inducer of a liquid rocket engine turbopump. Unsteady simulations were performed for the full 3D model of an inducer using a cavitation model. The inducer has been tested with water in the cavitation tunnel at JAXA-KSPL to examine suction performance and unsteady cavitation phenomena such as rotating cavitation and cavitation surge. Experiments were conducted under various flow conditions to examine a break-down point of the suction performance and unsteady cavitation phenomena. They have suggested that the casing geometry affected the onset of unsteady cavitation phenomena. Simulations were, therefore, performed for various cavitation numbers. The steady state was firstly calculated without a cavitation model, and then the unsteady calculation was done with the bubble two-phase flow model as a cavitation model. The effect of different model parameters on cavity structure was also examined. In the calculated results, it was clearly observed that the cavity structure grew on the blade surface and accompanied with vortices. These cavities showed dynamic change of their shapes as the rotation of the inducer. The calculated head coefficient showed decrease for small cavitation numbers with similar gradient to that observed in the experiment.

Commentary by Dr. Valentin Fuster
2005;():591-596. doi:10.1115/FEDSM2005-77430.

For experimental investigations of the thermodynamic effect on a cavitating inducer, it is nesessary to observe the cavitation. However, visualizations of the cavitation are not so easy in cryogenic flow. For this reason, we estimated the cavity region in liquid nitrogen based on measurements of the pressure fluctuation near the blade tip. In the present study, we focused on the length of the tip cavitation as a cavitation parameter. Comparison of the tip cavity length in liquid nitrogen (80 K) with that in cold water (296 K) allowed us to estimate the strength of the thermodynamic effect. The degree of thermodynamic effect was found to increase with an increase of the cavity length. The estimated temperature depression caused by vaporization increased rapidly when the cavity length extended over the throat. In addition, the estimated temperature inside the bubble nearly reached the temperature of the triple point when the pump performance deteriorated.

Topics: Nitrogen
Commentary by Dr. Valentin Fuster
2005;():597-602. doi:10.1115/FEDSM2005-77435.

Electro-Chemical Machining (ECM) is an advanced machining technology. It has been applied to highly specialized fields such as aerospace, aeronautics and medical industries. However, it still has some problems to be overcome. The efficient tool-design, electrolyte processing, and disposal of metal hydroxide sludge are the typical issues. To solve such problems, CFD is considered to be a powerful tool in the near future. However, the numerical method that can satisfactorily predict the flow has not been established because of the complex flow natures. In the present study, we investigate the modelling of the three-phase flow (i.e. fluid, hydrogen bubble and metal sludge) in ECM process. First, the developed code is applied to the two-dimensional channel configuration. The interactions among three-phases and the dissolved wall are simulated, to verify the modelling and to determine the model parameters, Next, the sinusoidal channel is machined by our code. It is confirmed that hydrogen bubbles in the separation region suppress the dissolution of the wall, and make the final wall shape be wavy. Through this study, it is exhibited that our developed model and code are sound and useful for simulating ECM process.

Commentary by Dr. Valentin Fuster
2005;():603-607. doi:10.1115/FEDSM2005-77448.

The two-dimensional test section of the ARL/Penn State 12-inch water tunnel has been modified to allow a wide range of dynamic tests using hydrofoils. Three examples of test configurations for the hydrofoil test facility are given. These include tests of a single conventional hydrofoil with non-sinusoidal deflections, tests with two hydrofoils for studying tip vortex interactions, and tests of base ventilated supercavitating hydrofoils. When testing a single hydrofoil, the angle of attack can be varied as a function of time using a cam drive system. Arc length Reynolds number of over 1 million based on a 1.5-inch chord are possible. Hydrofoil lift, drag and pitching moment can be measured during transient operation with and without cavitation. Tip vortex interaction studies have been performed by using a second hydrofoil mounted upstream of the primary test hydrofoil. This upstream hydrofoil is inclined to the tunnel wall so only the tip projects in front of the downstream hydrofoil. The upstream hydrofoil can be traversed across the test section to study the tip vortex interactions. Supercavitating hydrofoils have been tested by ventilating behind a wedge installed along the tunnel wall upstream of the hydrofoil. A full range of test instrumentation is used to support the studies, such as LDV, PIV, high speed video, and acoustic measurements.

Commentary by Dr. Valentin Fuster
2005;():609-616. doi:10.1115/FEDSM2005-77467.

A Large Eddy Simulation (LES) approach for cavitating flow, based on a virtual single-phase, fully compressible cavitation model which includes the effects of incondensable gas, has been shown to be capable of capturing the complex dynamical features of highly unsteady cavitating flows of two-dimensional hydrofoils. Here the LES results are compared to Time-Resolved Particle Image Velocimetry (TR-PIV) in the wake of a cavitating NACA 0015 hydrofoil, with particular attention to the predicted vortex shedding mechanisms. Despite some difficulty with obtaining vector fields from vortical clouds of vaporous-gaseous bubbles with cross-correlation techniques, the initial results seem promising in that they confirm the existence of a primary vortex pair (type A-B). In addition to TR-PIV, the cavitation cloud shedding was also documented with phase-locked, time-resolved photography and high speed volume-illuminated video, both with simultaneous imaging of side and plan views of the foil. All three experimental techniques confirm the need for fully three-dimensional simulations to properly describe the unsteady, three-dimensional cavitation cloud shedding mechanism.

Commentary by Dr. Valentin Fuster
2005;():617-622. doi:10.1115/FEDSM2005-77477.

In the present study, we have carried out an experimental investigation on the fluid-structure interaction caused by Karman vortices in the wake of a truncated 2D hydrofoil. The instrumentation involves a high frequency accelerometer and high speed visualisation. The mechanical response of the hydrofoil to the hydrodynamic excitation is monitored with the help of a portable digital vibrometer. Moreover, a specific optical device is developed to investigate the dynamic of the cavitating wake. The survey of the generation frequency of the Karman vortices with respect to the flow velocity reveals a Strouhal behaviour and three resonances of the hydrofoil. Out of hydro-elastic coupling conditions, the observation of the vortex structures reveals a strong 3D pattern despite the fact that the hydrofoil is 2D. The maximum fluid-structure interaction occurs for the torsional mode where lock-in is observed for upstream velocities ranging from 11 to 13 m/s. In this case, the vortices exhibit a 2D structure. The cavitation occurrence within the core of Karman vortices leads to a significant increase of their generation frequency. We have observed that hydrofoil resonance may be whether avoided or triggered by cavitation development. The study of the Karman vortices dynamic reveals that their advection velocity increases (4%) with the development of the wake cavitation meanwhile their streamwise spacing decreases.

Commentary by Dr. Valentin Fuster
2005;():623-629. doi:10.1115/FEDSM2005-77485.

A new unsteady, cavitation model for dense cloud cavitation is presented wherein the phase change process (bubble growth/collapse) is coupled to the acoustic propagation in a multi-phase fluid. This cavitation model predicts the number density and radius of bubbles in vapor clouds by tracking both the aggregate surface area and volume fraction of the cloud. Hence, formulations for the dynamics of individual bubbles (e.g. Rayleigh-Plesset equation) may be integrated within the macroscopic context of a dense vapor cloud i.e. a cloud that occupies a significant fraction of available volume and contains numerous bubbles. This formulation has been implemented within the CRUNCH CFD, which has a compressible “real” fluid formulation, a multi-element, unstructured grid framework, and has been validated extensively for liquid rocket turbopump inducers. Rigorous validation of the formulation is presented for various cases including unsteady simulations of a cavitating NACA0015 airfoil where the frequency of pressure fluctuations and time-averaged mean cavity lengths were compared with experimental data. The model also provides the spatial and temporal history of the bubble size distribution in the vapor clouds that are shed, an important physical parameter that is difficult to measure experimentally and is a significant advancement in the modeling of dense cloud cavitation.

Topics: Cavitation
Commentary by Dr. Valentin Fuster

Forum on Applications in CFD

2005;():631-635. doi:10.1115/FEDSM2005-77052.

In an attempt to explain the high loss of panels from the south face of the Vehicle Assembly Building (VAB) during Hurricane Frances, a three-dimensional computational fluid dynamics (3-D CFD) model was developed to simulate local velocity and pressure distributions resulting from such a storm. A preconditioned compressible Navier-Stokes flow solver1 was used to compute the flow field around the VAB complex, including the Launch Control Center, the Low and High Bays of the VAB, and several outbuildings in the immediate LC-39 area. The mapping of the forces and velocities on and along the affected faces of the VAB correlated surprisingly well with the extensive damage areas realized on both on the south face and on the southeast section of the roof. The model results were also consistent with the minimal damage seen on the east, north, and west faces of the structure.

Commentary by Dr. Valentin Fuster
2005;():637-640. doi:10.1115/FEDSM2005-77145.

In the present study criterion for local thermal equilibrium assumption is studied. It concerns with the fluid flow and heat transfer between two parallel plates filled with a saturated porous medium under non-equilibrium condition. A two-equation model is utilized to represent the fluid and solid energy transport. Numerical Finite Volume Method has been developed for solving coupled energy equations and the Non-Darcian effects are considered for description of momentum equation. The effects of suitable non dimensional parameters as Peclet number and conductivity ratio has been studied thoroughly. A suitable non dimensional equation proposed in wide range of Peclet number and conductivity ratio. This equation shows the temperature difference between solid and fluid phases.

Commentary by Dr. Valentin Fuster
2005;():641-646. doi:10.1115/FEDSM2005-77203.

Buoyancy-driven flow in a cavity can be observed in many application of thermal engineering, such as nuclear reactor insulation, ventilation of rooms, solar energy absorber, crystal growth, cooling of electronics chips, and cooling of steel or iron bars. Buoyancy-driven flow in a cavity has been studied using different solution methods. Validity of these numerical studies is performed by well-known benchmark problems. Differentially heated square cavity is one of the most widely used benchmark problems in convection-diffusion problems. In this study, the differentially heated cavity problem is modeled using control-volume method using different solution algorithms (QUICK and Upwind). The case matrix is formed (i.e. for different Ra number and for different mesh sizes) for this problem and the results are compared with the available solutions found in literature. The CPU time for the computations for the different cases with different solution algorithms are also given.

Commentary by Dr. Valentin Fuster
2005;():647-650. doi:10.1115/FEDSM2005-77222.

The present investigation deals with numerical prediction of airflow pattern in a room (enclosure) with a specific location of inlet and outlet with different values of Gr/Re2 . Two-dimensional, steady, incompressible, laminar flow under Boussinesq’s approximation has been considered. The velocity and temperature distributions in a room have been found by solving Navier Stokes equations and energy equation numerically by SIMPLE and SIMPLEC algorithms.

Commentary by Dr. Valentin Fuster
2005;():651-656. doi:10.1115/FEDSM2005-77225.

A 3-D full Navier-Stokes simulation of a large scale computing facility at NASA Ames Research center was carried out to assess the adequacy of the existing air handling and conditioning system. The flow simulation of this modern facility was modeled with a viscous, compressible flow solver code OVERFLOW-2 with low Mach number pre-conditioning. A script was created to automate geometry modeling, grid generation, and flow solver input preparation. A new set of air-conditioning boundary conditions was developed and added to the flow solver. Detailed flow visualization was performed to show temperature distribution, air-flow streamlines and velocities in the computer room.

Commentary by Dr. Valentin Fuster
2005;():657-662. doi:10.1115/FEDSM2005-77314.

The paper presents 2-D numerical simulations of laminar backward-facing step flow using the FemLab 3.1 modeling package. Results demonstrated that primary reattachment lengths predicted by FemLab were in close agreement with experimental data up to step Reynolds number Reh = 300. Also, dimensionless velocity profiles along the channel height calculated by FemLab were successfully compared with the experimental data.

Commentary by Dr. Valentin Fuster
2005;():663-667. doi:10.1115/FEDSM2005-77345.

Aerodynamic trade studies in support of an interdisciplinary research program for large ground based telescopes are addressed. Numerous CFD (Computational Fluid Dynamics) trade studies were carried out to help identify the initial critical configuration of the telescope. The highest pressures induced on the reflective surface of the telescope mirror in the critical configuration were used in structural analysis. A module that correlated disparate discretizations in structural and fluid analyses through common parent geometry was developed. This module mapped surface pressures from the CFD discretization to the structural discretization using a weighted average technique. Experimental validation of the CFD results was carried out in the University of Kansas subsonic wind tunnel. The results from the CFD analysis and the wind tunnel experiments were in close agreement, with the maximum variation of pressures being 1%–8%. The preliminary telescope configuration that induced the highest pressure on the reflective surface of the primary mirror was identified as one inclined at 60° from the vertical plane and facing the wind directly. An “open-air” CFD model was developed that simulated the observatory shut-off operating conditions of 15 m/s wind speed and a fail-safe operating condition of 50 m/s wind speed. Critical local total gage pressures were 165 Pa and 1400 Pa at 15 and 50 m/s wind speeds respectively.

Commentary by Dr. Valentin Fuster
2005;():669-676. doi:10.1115/FEDSM2005-77375.

Segregation in particulate multiphase flow with binary solid mixture has extensive applications in industrial separation processes. Up to now there have been few attempts towards numerical simulation of segregation in particulate multiphase flow with binary mixture due to complexity of the problem. In view of this, the primary objective of present work is to simulate the problem by computational fluid dynamics (CFD) and to validate by comparison with experimental measurements. Eulerian-Eulerian approach, incorporating the granular temperature, an essential ingredient in the solids pressure and solids viscosity formulation, was used to model the flow field of multiphase flow and was solved by Fluent 6.0. The CFD simulation results have been validated by experiments of liquid fluidization of binary solid mixtures. Validation results show that CFD simulation predict segregation and solid volume fraction profile precisely, and in addition, it can supply a more realistic prediction of other hydrodynamic features of the multiphase flow, such as velocity vector of all phases and pressure drop. The success of such CFD simulations opens doors for many potential studies.

Commentary by Dr. Valentin Fuster
2005;():677-681. doi:10.1115/FEDSM2005-77399.

We developed a voxel code using a finite difference method for unsteady three-dimensional thermal flows in a Cartesian grid system. This code enables us to predict flow fields around complicated geometries in a short pre-processing time. The code was used to predict flow fields around a flat plate that was heated to a constant temperature, and the results were within 20% of those obtained using analytical solutions. The code was also used to predict the flow field of a liquid crystal display (LCD) projector that has highly complicated internal structures such as a lamp, LCD panels, electric parts, etc.

Commentary by Dr. Valentin Fuster
2005;():683-688. doi:10.1115/FEDSM2005-77421.

This paper presents computational results for two DES (Detached Eddy Simulation), one hybrid RANS (Reynolds Averaged Navier-Stokes)/ LES (Large Eddy Simulation) and some preliminary results from PANS (Partially Averaged Navier-Stokes) turbulence for simulation of unsteady separated turbulent flows. The models are implemented in a full 3-D Navier Stokes solver and are based on the two-equation k-ε model. The formulations of each model are presented and results are analyzed for subsonic flow over a Backward Facing Step (BFS). Simulations are carried out using a 3rd order Roe scheme. A comparative assessment is made between the predictions from the DES, hybrid and PANS models. The predicted results are compared with the available experimental data for skin-friction coefficient, and different turbulent quantities. The three-dimensionality of the flow field and the separated fine scale structures are presented through the Q iso-surfaces.

Commentary by Dr. Valentin Fuster
2005;():689-693. doi:10.1115/FEDSM2005-77425.

This paper describes the application of computational fluid dynamics (CFD) to fan design as part of the PAX Scientific product development cycle. Six-inch fans are investigated using a simulation configuration designed to match in-house testing facilities. Grid generation and flow simulations are carried out using a commercial CFD software (Fluent). The segregated solver is used to run steady state RANS simulations with a Realizable k-ε turbulence scheme. Performance predictions (flow and pressure generated, power required) are in agreement with experimental testing values, with a margin of error ranging from less than 1% to less than 10%.

Commentary by Dr. Valentin Fuster
2005;():695-700. doi:10.1115/FEDSM2005-77432.

Turbulent flow over a rough surface with suction or blowing is a common fluid mechanics problem that has many practical applications including pulp screening. The present study is motivated by the optimization of aperture geometry in pulp screens used in the pulp and paper industry to separate unwanted contaminants from pulp fibres. In these devices, a dilute suspension of fibres is forced through fine slots to remove the oversized contaminants. To better understand the complex hydrodynamics at the critical region near the screen surface, a Computational Fluid Dynamics study has been conducted to examine how the geometry of the aperture entry, characterized by contour height and wire width, affect the details of the flow field. The results indicate that the flow is dominated by a separation vortex that covers the aperture. For low contour heights the flow is accelerated upstream of the aperture due to the high vorticity at the aperture entry and as profile height increases the turbulence intensity at the screen cylinder surface increases. Although the wire width has less of a impact on the flow field than contour height, the results show that decreasing wire width increases turbulence intensity at the surface due to a increase in apparent roughness of the cylinder.

Commentary by Dr. Valentin Fuster
2005;():701-706. doi:10.1115/FEDSM2005-77433.

In this study the performance of supersonic and hypersonic impactors under various operating conditions was analyzed using a computer simulation model. The study was focused on the effect of the nozzle upstream condition on the performance of the supersonic and hypersonic impactors. In our earlier work, the computational domain covered downstream of the nozzle with a sonic boundary condition at the inlet. In the present study, the computational domain included the upstream nozzle where the flow and particles enter with at low velocities. Axisymmetric forms of the compressible Navier-Stokes and energy equations were solved and the gas flow and thermal condition in the impactor were for evaluated. A Lagrangian particle trajectory analysis procedure was used and the deposition rates of different size particles under various operating conditions were studied. For dilute particle concentrations, one-way interaction was assumed and the effect of particles on gas flow field was ignored. The importance of drag and Brownian forces on particle motions in supersonic/hypersonic impactors was analyzed. Sensitivity of the simulation results to the use of different expressions for the drag force was also examined. It was shown that when the upstream nozzle is included in the computational model, the Stokes-Cunningham drag with variable correction coefficient and a constant Cunningham correction factor based on stagnation point properties lead to the same results. Thus these drag laws are most suitable for computer simulation studies of nano-particles in supersonic/hypersonic impactors. The computer simulation results were shown to compare favorably with the experimental data.

Commentary by Dr. Valentin Fuster
2005;():707-712. doi:10.1115/FEDSM2005-77434.

The case of a supersonic turbulent flows with Mach number 2.5 and Reynolds number 1.23×106 based on the diameter of after body, around a body with incidence angles of 14° was studied. The nose length was 3 times the diameter with a third degree polynomial variation, and total length of the body was 13 diameters. Reynolds Averaged Navier-Stokes Equation was solved using central differencing scheme. The Reynolds Stress Model was used to account for the effect of turbulence on the flow field. The experimental data consist of surface pressure measurement at six axial locations. The pressure distributions were compared with the experimental data and the computer simulation results using Baldwin-Lumax and k-ε models. RSM results were found to show good agreement with the experimental data, while the Baldwin-Lumax model predictions deviated from the experimental data at the leeward on the after body because of a large cross-flow separation. The cross-sectional Mach number contours were also presented. It was shown that in addition to the outer shock, a cross-flow shock wave was also present in the flow region.

Commentary by Dr. Valentin Fuster
2005;():713-715. doi:10.1115/FEDSM2005-77439.

At the last minute, a separation vessel is adapted in response to an unexpected alteration in its original design. While vessel internals are undergoing fabrication in the shop, a revised external piping layout necessitates changes to the overall vessel design. This vessel is compact and its design takes full advantage of its exact geometry. Specifically, the inlet and outlet orientations, to be modified, are critical to the original design. To keep the project on schedule, a new design, created and validated by an expedited computational fluid dynamics (CFD) study, is proposed for installation in 2005. The steps taken to validate the design are described in this paper.

Commentary by Dr. Valentin Fuster
2005;():717-720. doi:10.1115/FEDSM2005-77441.

The work to be presented herein is a Computational Fluid Dynamics analysis of flow over a 15-degree angle double wedge for a compressible air, with Mach number of 2.95. The problem to be solved involves formation of shock waves, expansion fans and slip surfaces, so that the general characteristics of supersonic flow are explored through this problem. Shock waves and slip surfaces are discontinuities in fluid mechanics problems. It is essential to evaluate the ability of numerical technique that can solve problems in which shocks and contact surfaces occur. In particular it is necessary to understand the details of developing a mesh that will allow resolution of these discontinuities. Results including contour plots of pressure, temperature, density and Mach number will show that CFD is capable of predicting accurate results and is also able to capture the discontinuities in the flow, e.g., the oblique shock waves and the slip surfaces. The global comparison of some parameters between the numerical and the analytical values show a good agreement.

Commentary by Dr. Valentin Fuster
2005;():721-728. doi:10.1115/FEDSM2005-77443.

In this study, the influence of surface roughness in the prediction of the mean flow and turbulent properties of a high-speed supersonic (M = 2.9, Re/m = 2.0e7) turbulent boundary layer flow over a flat plate is performed using the k-ω and the stress-ω models. Six wall topologies, including a smooth and five rough surfaces consisting of three random sand-grain plates and two uniformly machined plates were tested. Experimental data are available for these configurations. It is observed that, for smooth surface, both k-ω and stress-ω models perform remarkably well in predicting the mean flow and turbulent quantities in supersonic flow. For rough surfaces, both models matched the experimental data profiles fairly well for lower values of the roughness height. Overall, the k-ω model performed better than the stress-ω model. The stress-ω model did not show any strong advantages to make up for the extra computational cost associated with a Reynolds stress model. The simulation results indicated that the prescription for the surface boundary conditions for ω in both models, especially for the stress-ω model, need to be refined encountering high roughness numbers and reconsidered to include the geometric factor.

Commentary by Dr. Valentin Fuster
2005;():729-736. doi:10.1115/FEDSM2005-77445.

This work deals with a numerical simulation developed to predict the characteristic cooling times of a low-thermal diffusivity fuel-oil confined in the tanks of a wrecked ship. A typical scenario has been introduced, through the definition of tanks geometries, physical boundary conditions (deep sea temperatures) and reological properties of the fuel-oil. The fluidynamic behaviour of the oil (forced convection) inside the tanks, as well as the heat exchange with surrounding sea water has been simulated throughout a commercial code, FLUENT, that solves directly the Navier-Stokes set of equations, including energy one. The purpose is focused on the prediction of both spatial and temporal evolution of the fuel-oil characteristic temperature inside the tanks. The final objective is placed on the determination of the deadline in which asymptotic temperature curve of the fuel-oil converges to deep sea thermal conditions. Inspectional analysis is also outlined, as a powerful tool to predict an order of magnitude in the cooling process.

Commentary by Dr. Valentin Fuster
2005;():737-740. doi:10.1115/FEDSM2005-77449.

A selective catalytic reduction (SCR) system, when designed for a simple cycle turbine, presents a significant calculation and modeling challenge due to its compact design and stringent performance requirements. In particular, uniform flue gas velocity profiles, required by environmental catalysts installed in the ductwork of this system, must be met. Custom flow devices optimized for the turbine. SCR system and ductwork are required. Cold flow and computational fluid dynamics (CFD) modeling are employed to design flow devices that provide adequate velocity profiles. The purpose of this paper is to present (1) steps taken to optimize the ductwork internals and (2) measured and calculated velocity profiles.

Commentary by Dr. Valentin Fuster
2005;():741-746. doi:10.1115/FEDSM2005-77450.

A new law of the wall accounting for curvature effects for swirling axial flows was derived. The curvature influence on turbulent mixing lengths in both axial and tangential directions was examined theoretically using Reynolds Stress Model. For equilibrium flows and weak curvature, identical mixing lengths were derived for both directions. In addition, the effect of shear stress ratio on the near-wall velocities was explored systematically. It was found that the curvature effect in swirling axial flows was suppressed by a factor of 1/(1+σw 2 ) compared to that in curved channel flows. Considerably improved agreement with measurements was obtained for the new curvature law compared to the classical logarithmic law.

Commentary by Dr. Valentin Fuster
2005;():747-752. doi:10.1115/FEDSM2005-77463.

Numerical simulations were conducted to investigate the feasibility of predicting near field concentrations of a tracer gas within and above forest canopies. The current research is geared towards providing forest managers with a tool for developing anti-aggregation techniques to control the bark beetle. Several field experiments have been conducted in different forest canopies linking tracer gas concentration fields with meteorological and canopy parameters. Field experiment results are site and situation specific. Numerical simulations are far less expensive and allow for variation in virtually all flow parameters such as atmospheric stability, wind speed and direction, and turbulence intensity. As a first step, a CFD simulation has been used to study dispersion in a generic lodgepole pine forest canopy based on leaf area index (LAI) and stem density. Steady Reynolds Averaged Navier Stokes (RANS) solutions were computed using the k-ε and Reynolds Stress Model (RSM) turbulence closure models. These solutions provide insight into in-canopy dispersion; however they do not fully capture the dynamics of the flow. The current work uses large eddy simulation (LES). LES resolves large flow dominated eddies while modeling smaller eddies using a sub grid scale model. Unsteady LES, can be used to capture the dynamics of flow within a canopy, including large rolling eddies above the canopy, bursting and sweeping within the canopy, multiple shear layers, and drainage flows.

Commentary by Dr. Valentin Fuster

Forum on Microfluidic Devices and Systems

2005;():753-760. doi:10.1115/FEDSM2005-77378.

This paper describes an application of a general purpose computer program, GFSSP (Generalized Fluid System Simulation Program) for calculating flow distribution in a network of micro-channels. GFSSP employs a finite volume formulation of mass and momentum conservation equations in a network consisting of nodes and branches. Mass conservation equation is solved for pressures at the nodes while the momentum conservation equation is solved at the branches to calculate flowrate. The system of equations describing the fluid network is solved by a numerical method that is a combination of the Newton-Raphson and successive substitution methods. The numerical results have been compared with test data and detailed CFD (computational Fluid Dynamics) calculations. The agreement between test data and predictions is satisfactory. The discrepancies between the predictions and test data can be attributed to the frictional correlation which does not include the effect of surface tension or electro-kinetic effect.

Commentary by Dr. Valentin Fuster
2005;():761-770. doi:10.1115/FEDSM2005-77406.

Hydrodynamic cavitation, the explosive growth and catastrophic collapse of vapor bubbles, has immense impact on the design and performance of hydraulic machinery in the macro world. However, cavitation in high-speed microfluidic systems has received scarce attention and hardly been reported. This article reports the presence of hydrodynamic cavitation in the flow of de-ionized water through 11.5–40micron wide rectangular slot micro-orifices entrenched inside 100–200micron wide microchannels. Pioneering experimental investigations on hydrodynamic cavitation in rudimentary microfluidic configurations such as slot micro-orifices has been presented and unique cavitating flow patterns have been identified. Assorted cavitating (two-phase) flow patterns including incipient, choking and supercavitation have been detected. Designers of high-velocity microfluidic systems, especially Power-MEMS devices, need to be aware of the deleterious effects of cavitation as it can significantly affect device performance. The effects of micro-orifice and microchannel size on cavitation have been discussed and results indicate the existence of strong scale effects. Incipient and choking cavitation numbers are observed to increase with increasing micro-orifice size, while the orifice discharge coefficient plummets once cavitation activity erupts. In addition, inlet pressure effects on several cavitation parameters have been discussed and compared with established macro-scale results. The cavitating flow patterns encountered are significantly influenced by the micro-orifice and microchannel size. Flow rate choking occurs irrespective of the inlet pressures and is a direct consequence of cavitation inside the micro-orifice. Cavitation hysteresis is observed but its effects are more marked for the smallest micro-orifice.

Commentary by Dr. Valentin Fuster

Forum on Applications of Automotive Flows

2005;():771-776. doi:10.1115/FEDSM2005-77102.

A novel method for simulating the relative motions of the wheels and moving ground for road vehicle aerodynamics is presented. The method revisits an old concept where two identical vehicles are used and positioned so that they are mirror images, with the ground being represented by the horizontal plane of symmetry. The method involves double symmetry, where two half models (e.g. a car split down the vertical centerline) contact at the rotating wheel contact patches and the resulting (opened) vehicle halves lie on a reflection plane. This can either be the tunnel floor or the equivalent CFD plane. For some forms of physical testing this offers advantages (such as easy access to wheel cavities and requiring only one vehicle) but sealing the gap between the tunnel floor and the vehicle halves can interfere with the force balance accuracy and problems can arise with time-varying flows crossing the time averaged zero flow boundary. This paper describes the concept and CFD and model-scale EFD evaluations which were found to compare well.

Commentary by Dr. Valentin Fuster
2005;():777-785. doi:10.1115/FEDSM2005-77324.

An unsteady computational fluid dynamics simulation of the viscous-turbulent flow around a tractor-trailer has been done using FLUENT’s realizable k-epsilon turbulence model and is supported by experimental validation. The primary objective was to compute a time-dependent solution of the flow around a tractor-trailer in a virtual wind tunnel and study the pressures on the floor.

Commentary by Dr. Valentin Fuster
2005;():787-791. doi:10.1115/FEDSM2005-77461.

Understanding the disintegration mechanism, spray penetration, and spray motion is of great importance in the design of a high quality diesel engine. The atomization process that a liquid would undergo as it is injected into a high-temperature, high-pressure air, is investigated in this work. The purpose of this study is to gain further insight into the atomization mechanism, the variation over time in droplet size distribution and spray penetration. This is done based on the effects of chamber pressure, injection pressure, and type of fuel. A laser diffraction method is used to determine droplet mean diameters, single injection with synchronized time mechanism allowed the time dependent studies. Obscuration signals are obtained through a digital oscilloscope from which arrival time of spray can be measured. A spray penetration correlation is reported.

Topics: Drops , Sprays
Commentary by Dr. Valentin Fuster

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