0


RESEARCH PAPERS

J. Fluids Eng. 1999;121(3):517-525. doi:10.1115/1.2823499.

The novel quintuple hot-wire measurement technique was used to perform detailed measurements of the mean velocities and Reynolds stresses in an isothermal model combustion chamber at two different levels of swirl. The measured flow quantities are analyzed and described in detail where the emphasis is put on typical swirl-related effects as well as the interaction of rotation and turbulence dynamics. The results provide a well-documented data base for the development and validation of turbulence closures. They also serve to improve understanding of specific characteristics of swirl flows.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):526-532. doi:10.1115/1.2823500.

Turbomachinery flows can be extremely difficult to predict, due to a multitude of effects, including interacting strain rates, compressibility, and rotation. The primary objective of this investigation was to study the influence of multiple strain rates (favorable streamwise pressure gradient combined with radial pressure gradient due to convex curvature) on the structure of the turbulent boundary layer. The emphasis was on the initial region of curvature, which is relevant to the leading edge of a stator vane, for example. In order to gain better insight into the dynamics of complex turbulent boundary layers, detailed velocity measurements were made in a low-speed water tunnel using a two-component laser Doppler velocimeter. The mean and fluctuating velocity profiles showed that the influence of the strong favorable pressure augmented the stabilizing effects of convex curvature. The trends exhibited by the primary Reynolds shear stress followed those of the mean turbulent bursting frequency, i.e., a decrease in the bursting frequency coincided with a reduction of the peak Reynolds shear stress. It was found that the effects of these two strain rates were not superposable, or additive in any simple manner. Thus, the dynamics of the large energy-containing eddies and their interaction with the turbulence production mechanisms must be considered for modeling turbulent flows with multiple strain rates.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):533-540. doi:10.1115/1.2823501.

Drag reduction was investigated for the combined system of polymer additives and a riblet pipe. The riblet grooves were V-shaped, the spacing of which was 1.3 mm and the height of which was 1.01 mm. For higher h+ , a triangular riblet system including other geometries increases the drag to levels similar to those of normal transient roughness. This drag increase was generally given as a function of h+ . The polymer additives were Aronfloc N-110 and Separan AP-30. The critical shear stress τ*, at which N-110 started the drag reduction, was approximately eight times higher than τ* for AP-30. In the combined system, the synergistic drag reduction for higher h+ was discussed under the assumption that the additives suppressed the drag increase resulting from riblets. Since the additives thicken a wall layer covering the region from a viscous sublayer to a buffer layer, the relative height of h to this wall layer thickness is lowered. In addition, the flow enhancement due to additives relatively suppresses the riblet-induced drag increase. The analysis based on velocity profiles indicated that these effects can produce synergistic drag reduction for higher h+ .

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):541-547. doi:10.1115/1.2823502.

Laminar drag reduction has been shown for the flow of a Newtonian fluid in the space between two vertical coaxial cylinders. Experiments were carried out to measure the torque of a bob with a highly water-repellent wall to clarify the effect of the contact surface of the bob on the flow behavior. The basic material of the highly water-repellent wall is fluorine alkane modified acrylic resin with added hydrophobic silica, and the contact angle of the wall is about 150 degree. The radius rations of the bob were 0.932 and 0.676. Test fluids were Newtonian aqueous solutions of 60, 70, and 80 wt% glycerin and polymer solutions. The maximum drag reduction ratio was about 12% for 80 wt% glycerin solution at a radius ratio of 0.932. The moment coefficient of the coaxial cylinder in Newtonian fluids was analyzed for fluid slip, and it was shown that the analytical results agreed well with the experimental data. For the case of non-Newtonian fluids, the fluid slip velocity of polymer solutions is not proportional to the shear stress and the relationship is approximated by power-law equations.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):548-554. doi:10.1115/1.2823503.

A linear stability analysis has been carried out for hydromagnetic current-induced flow. A viscous electrically conducting fluid between concentric cylinders is driven electromagnetically by the interaction of a superimposed radial current and a uniform axial magnetic field. The assumption of small-gap approximation is made and the governing equations with respect to nonaxisymmetric disturbances are derived and solved by a direct numerical procedure. Both of the two different types of boundary conditions, namely ideally conducting and weakly conducting walls, are considered. For 0 ≤ Q ≤ 5000, where Q is the Hartmann number, which represents the strength of magnetic field in the axial direction, it is found that the instability sets in as a steady secondary flow for the case of weakly conducting walls but not for ideally conducting walls. For ideally conducting walls, it is demonstrated that the onset mode is due to nonaxisymmetric rather than axisymmetric disturbances as Q exceeds a certain critical value. The transition of the onset of instability from axisymmetric modes to nonaxisymmetric modes is discussed in detail and the possibility of axisymmetric oscillatory modes is examined. The values of the radial current density required for the appearance of secondary flow are also determined. Furthermore, the predictions of present numerical results are found to be in agreement with previous experimental studies.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):555-559. doi:10.1115/1.2823504.

An experimental verification is presented for the theoretical vortex trapping concept. A suction orifice located along one wall of a water channel test section was used to simulate a point sink to trap spanwise vortices downstream of a backward-facing step and between two parallel fences. Results from the backward-facing step geometry indicated an increase in the sink strength required to hold a vortex as the sink is positioned closer to the step, closely following previous theoretical predictions made using conformal mapping. The experimental data also showed reasonable agreement with the theoretical position for optimum vortex trapping. Flow visualization has shown a three-dimensional cross-stream effect due to bending of the forced-vortex core by suction. Results from the dual-fence geometry, on the other hand, verified the ability to use a lower level of suction for vortex trapping when compared with the backward-facing step.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):560-567. doi:10.1115/1.2823505.

The flow in a torsional shear cell is investigated with the purpose of finding a measure for the wall effect due to strongly nonuniform flow in the vicinity of the edge of the top platen. Various laminar flow problems are analysed that are relevant to this set-up. These include pure shearing flow of a single fluid in a both an infinite and finite cell, as well as pure shear of a two-fluid system in a finite cell. For pure shearing flow it is found that the extent of the wall effect is of the order of magnitude of the depth of the fluid layer. For piston flow the wall effect is entirely determined by boundary conditions at the bottom of the cell.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):568-573. doi:10.1115/1.2823506.

A significant effect of permeability variations on the three-dimensional fluid flow in a heterogeneous porous channel subject to rotation is presented. The results of a numerical solution to the governing equations confirm for the more general case the conclusions from earlier analytical investigations, which suggest that permeability functions be classified corresponding to whether their variation is monotonic or not, and to whether their vertical gradient is positive or not. Unicellular and multiple vortex solutions are obtained for the secondary flow in the plane perpendicular to the imposed axial flow, while their direction is dictated by the corresponding class of permeability function as applicable. The impact of rotation on the imposed axial flow is shown to be significant as well, leading to different axial flow fields depending again on the class of permeability function used. In particular, the rotation impacts significantly in creating axial flow deficiencies in some regions on the cross section.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):574-579. doi:10.1115/1.2823507.

Incompressible, steady and pulsatile flows in axisymmetric sudden expansions with diameter ratios of 1:2.25 and 1:2.00 have been simulated numerically over the ranges of time-averaged bulk Reynolds number 0.1 ≤ Re ≤ 400 and Womersley number 0.1 ≤ W ≤ 50. For steady flow, the calculated recirculation zone length increased linearly with an increase in Re, in good agreement with earlier experiments. For pulsatile flows, particularly at higher values of W, the recirculation zone length correlated strongly with the acceleration of the flow and not with the instantaneous Reynolds number; it increased during the deceleration phase and decreased during the acceleration phase. The computed mean velocity and reattachment length were in general agreement with published experimental data. At relatively low W, the computed near-wall, reverse flow region extended along the full domain over part of the cycle, similarly to that in the experiments. At low values of W, the vortex rings created at the expansion remained attached and oscillated back and forth; for an intermediate range of W, they detached and moved downstream; at relatively high W, these vortices became, once more, attached.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):580-587. doi:10.1115/1.2823508.

Turbulent flowfields resulting from an oblique jet injecting from a rectangular side-inlet duct into a rectangular main duct with an aspect ratio 3.75 without a forced crossflow are presented in terms of laser-Doppler velocimetry measurements. The main focuses are the effects of the side-jet angle (θ) and side-jet flow rate (Qs ) on the mass entrainment upstream of side-jet port and the flowfield in the rectangular duct. The side-inlet angles investigated were 30, 45, and 60 deg and Reynolds numbers based on the air density, rectangular duct height and bulk mean velocity were in the range of 7.1 × 103 to 3.6 × 104 corresponding to Qs values of 1 × 103 to 5 × 103 L/min. The present study suggests the presence of a critical side-injection angle θc (30 ≤ θc ≤ 45 deg) above which a recirculation zone appears in the rectangular duct, whereas below which the recirculation zone is absent. For the more tangential angle (θ = 30 deg), almost as much fluid is entrained into the main duct as was injected from the side jet. The mean flow field in the rectangular duct is found to be a weak function of the Reynolds-number for the range of Qs investigated. In addition, a simple linear correlation between the mass entrainment upstream of the side-jet port and side-injection angle is obtained. Complementary flow visualizations and numerical computations with an algebraic Reynolds stress model were also performed. The discrepancies between measured and computed results are documented.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):588-595. doi:10.1115/1.2823509.

The application of LDA to a transient 1/3 scale (500 mm wide) water model of a mould, typical of steel thin slab casting, is presented. The characteristics of a crossflow, associated with the oscillating jet emerging from a submerged nozzle (internal diameter 33 mm), were analyzed for a range of casting rates (0–2 m/min), nozzle submergences (20–120 mm) and nozzle-mould wall gap widths (0–21 mm). The frequency of oscillation was found to be primarily dependent on the casting rate of the system, independent of nozzle submergence or gap width, whereas the RMS crossflow velocity depended on all three parameters. Additional crossflow was also observed past the jet below the nozzle exit and this allowed the jet to oscillate even with zero gap width.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):596-604. doi:10.1115/1.2823510.

The influence of convex curvature on the mean and turbulent characteristics of three-dimensional wall jet generated from a circular orifice geometry is reported in this paper. Mild, moderate, and strong curvature are considered. Detailed results on a plane surface are also obtained for comparison purposes. Among the mean properties, the decay rate of maximum velocity and the growth of length scales are significantly altered due to curvature effects. The turbulent components, both turbulent normal and shear stresses, show an increase with curvature parameter in a direction normal to the curved surfaces; however, there is very little change in the spanwise direction.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):605-613. doi:10.1115/1.2823512.

This paper describes the use of a mechanical pectoral fin as a new device for maneuvering and stabilizing an underwater vehicle. The mechanical pectoral fin consists of two servo-motors generating a feathering motion and a rowing motion. The hydrodynamic characteristics of the device were analyzed experimentally and theoretically. The mechanical pectoral fin generates a thrust force in a range of phase differences between both motions. The unsteady vortex-lattice method, which takes into account the effects of viscosity, reasonably expresses the unsteady forces acting on the mechanical pectoral fin.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):614-620. doi:10.1115/1.2823513.

Automotive torque converters have recently been designed with an increasingly narrower profile for the purpose of achieving a smaller axial size, which also translates into weight savings. Four torque converters with different flatness ratios were manufactured and tested in order to evaluate the change in their overall performance, including efficiency, stall torque ratio and torque transmission capacity. The experimental results show that the overall performance deteriorates when the flatness ratio is reduced to less than about 0.2. The internal flow characteristics of the torque converters were also investigated by numerical analysis using a CFD code. The computational results indicate that the main cause of this performance deterioration is a reduction in pump efficiency, which is attributed to increases in shock loss in the inlet region, separation loss in the fore half region, and friction loss in the exit region.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):621-626. doi:10.1115/1.2823514.

An experimental investigation is presented regarding the unsteady pressure field within a high specific speed centrifugal pump impeller (ωs = 1.7) which operated in a double spiral volute. For this, twenty-five piezoresistive pressure transducers were mounted within a single blade passage and sampled in the rotating impeller frame with a telemetry system. The influence of varying volume flux on the pressure transducers was evaluated in terms of pressure fluctuation magnitudes and phase differences. The magnitude information reveals that the pressure fluctuations from the impeller-volute interaction grew as the volume flux became further removed from the best efficiency point and as the trailing edge of the impeller blade was approached. These fluctuations reached 35% of the pump head in deep part load. The upstream influence of the volute steady pressure field dominates the unsteady pressure field within the impeller at all off design load points. Acquired signal phase information permits the identification of the pressure field unsteadiness within the impeller passage as fundamentally synchronized simultaneously with the volute tongue passing frequency. Special emphasis was placed on the volume flux regime where the pump and impeller pressure discharge characteristic undergo hysteresis, as impeller inlet and outlet recirculation commence and cease. A synthesis of the rotating transducers was performed to obtain unsteady blade loading parameters. The value of the unsteady lift coefficient varies on the order of 200% for a single blade in part load operation (at 45% bep), an abrupt fluctuation occurring as the fore running blade suction side passes a volute tongue. The unsteady moment coefficient and center of pressure are also shown to vary significantly during the impeller-volute tongue interaction.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):627-632. doi:10.1115/1.2823515.

Hysteresis in a pump characteristic results from instability phenomena involving complex three dimensional flow with recirculation. The unsteady flow field on the top and bottom branches of a hysteresis loop in a high specific speed (ωs = 1.7) centrifugal pump characteristic was experimentally evaluated. A hypothesis for recirculation zones and prerotation as power dissipaters is proposed for explaining the discrepancy in the pressure and shaft power hysteresis. The experimental investigation was performed in both the rotating and stationary frame. In the rotating frame 25 miniature pressure transducers mounted in an impeller blade passage were sampled with a telemetry system. In the stationary frame a fast response probe was implemented. The changing impeller flow field manifested itself between the two branches of the hysteresis with increasing stochastic pressure fluctuations. Using this information the position, size, and strength of the impeller recirculation was quantitatively determined. Theoretically the rate of change of useful hydraulic power in the hysteresis regime during transient pump operation was found to be a function of throttling rate. Quasi-steady behavior existed for slow throttling, |dφ/dt| < 0.005 s−1 . A second-order nonlinear dependence on the throttle rate was determined for the change of useful flow power during the commencement/cessation of the impeller recirculation.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):633-637. doi:10.1115/1.2823516.

Unsteady forces generated by fluid flow through the impeller shroud leakage path of a centrifugal pump were investigated. Different pump shroud geometries were compared, and the effect of leakage path inlet swirl (pump discharge swirl) on the rotordynamic forces was examined for various ratios of fluid throughflow velocity to impeller tip speed. A short axial length leakage path reduced the measured forces, while curvature appeared to increase the destabilizing forces when inlet swirl was present. It was observed that changing the inlet swirl velocity does not appear to significantly affect the measured forces for a given leakage flow coefficient, but any nonzero inlet swirl is destabilizing when compared to cases with no inlet swirl.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):638-645. doi:10.1115/1.2823517.

The uniform flow past a rotating marine propeller was studied using incompressible Reynolds-averaged Navier-Stokes computations with the Baldwin-Barth turbulence model. Extensive comparison with the experimental data was made to validate the numerical results. The general characteristics of the propeller flow were well predicted. The current numerical method, however, produced an overly diffusive and dissipative tip vortex core. Modification of the Baldwin-Barth model to better predict the Reynolds stress measurements also improved the prediction of the mean velocity field. A modified tip geometry was also tested to show that an appropriate cross section design can delay cavitation inception in the tip vortex without reducing the propeller performance.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):646-650. doi:10.1115/1.2823518.

A study is made on the pressure surges on a fluid system with air valves. The effects of the presence and distribution of air on the system response are studied. Several distinct pressure transient characteristics were observed through this investigation. Investigations showed that air valves with high inflow characteristics installed at peak locations of a fluid system with entrapped air may reduce the magnitude of the extreme negative pressure surges. However, for near zero air entrainment levels, air valves with higher outflow characteristics tend to result in higher positive pressure surges. The effectiveness of the air valves installed for the purpose of surge protection depends not only on the physical configuration of the fluid system, the physical properties of the pipeline and the fluid, but significantly also on the characteristics of the air valves and the distribution of air in the system. The results of the investigations in the present work were confirmed through field observations and measurements.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):651-655. doi:10.1115/1.2823519.

Nonlinear thin film rupture has been analyzed by investigating the stability of films under the influence of a nonuniform electrostatic field to finite amplitude disturbances. The dynamics of the liquid film is formulated using the Navier-Stokes equations including a body force term due to van der Waals attractions. The effect of the electric field is included in the analysis only in the boundary condition at the liquid vapor interface. The governing equation was solved by finite difference method as part of an initial value problem for spatial periodic boundary conditions. The electric field stabilizes the film and increases the time to rupture when a long wavelength perturbation is introduced.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):658-664. doi:10.1115/1.2823520.

A model is developed describing a novel ink jet printer driven by a piezoelectric component used to eject a fluid droplet through a rubber valve. The model is analyzed to address specific printer design issues. Rapid droplet production and efficient conversion of piezoelectric energy to droplet kinetic energy are ensured by suitable choices of the ejector geometry and the voltage step used to produce the droplet. A parameter regime is found in which a resonance prevents the valve from closing properly, and this particular regime must be avoided for correct printer operation. By choosing a suitable voltage signal after the production of an ink droplet, the device is returned rapidly to its initial or quiescent state.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):665-672. doi:10.1115/1.2823521.

Numerical simulation of the internal and external flow fields of a liquid drop moving in the surrounding gas are considered. The present work is concerned with the time accurate numerical solution of a two phase flow field at the low Mach number limit with an appropriate volume tracking method to capture motion and deformation of a liquid drop. In particular, deformation of a liquid drop moving with a coflowing gas stream in a zero gravity field is simulated. The effects of the gas flow Reynolds number and drop Weber number on the deformation dynamics of the drop have been investigated. There appears to be a critical gas stream Reynolds number, at moderate drop Weber numbers, below which the coflowing drop takes on an oblate cap shape and above which it forms an arrow head shape. It has been shown that an observer moving with the average velocity of the liquid drop sees interesting recirculatory flow patterns inside the drop.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):673-677. doi:10.1115/1.2823522.

Gravity-driven flow over a round-crested weir is analyzed for viscous flow. An equation for the entire flow profile is obtained by simplifying the equations for slowly varying film thickness, assuming a velocity profile, and integrating across the film. Solution of the resulting first order, ordinary differential equation requires a boundary condition generated at a critical point of the flow, beyond which waves cannot propagate upstream. Results for the relationship between head and flow rate are consolidated on a dimensionless master curve represented by an empirical equation.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):678-683. doi:10.1115/1.2823523.

Effects of geometry on the flux of vorticity from a free surface are discussed. Special attention is paid to situations where curvature-dependent contributions to the vorticity flux can be neglected. The geometry of vortex lines embedded in the surface is discussed in this context. These results show that vortex lines can be straight and geometry-induced vorticity flux is produced; conversely vortex lines can be curved and no geometry-induced vorticity flux is produced. A convenient method for assessing vorticity flux from a steady surface based on Gaussian curvature is derived.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):684-689. doi:10.1115/1.2823524.

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

Commentary by Dr. Valentin Fuster

ERRATA

TECHNICAL BRIEFS

J. Fluids Eng. 1999;121(3):690-693. doi:10.1115/1.2823525.
Abstract
Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1999;121(3):693-696. doi:10.1115/1.2823526.
Abstract
Topics: Vortices , Geometry
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