0


REPORTS

J. Fluids Eng. 1981;103(1):6-7. doi:10.1115/1.3240783.
FREE TO VIEW
Abstract
Commentary by Dr. Valentin Fuster

RESEARCH PAPERS

J. Fluids Eng. 1981;103(1):8-13. doi:10.1115/1.3240790.

Several devices have been developed or adapted to the use of measuring “nucleating sources” in a liquid including acoustic methods, direct observation (e.g. microscopy and holography), single particle counters, light scattering and light extinction. With experience two techniques have become more favored than the others, namely: holography and light scattering. However, with the exception of Peterson et al’s study no attempt has been made to simultaneously compare these two techniques under actual test conditions so that some confidence may be gained in deducing the number and type of nuclei from the record produced by either method. The purpose of the present paper is then to report and compare some nuclei populations which were recorded simultaneously by both methods during a series of cavitation inception tests. Measurements of nuclei populations were made in two facilities located at the California Institute of Technology, the Low Turbulence Water Tunnel (LTWT) and the High Speed Water Tunnel (HSWT) which are known to have substantially different nuclei distributions. In each facility nuclei populations were recorded at various combinations of tunnel velocity, pressure and pressure-time histories in an effort to produce different distributions and also to enable one to deduce from the laser scattering results the composition of the nuclei, i.e., solid particulates or bubbles. This deduction could then be checked with the holographic results. Simultaneous nuclei population measurements were recorded in the LTWT, whereas measurements of nuclei in the HSWT were obtained at separate times approximately eleven months apart. Since both methods have different characteristic size detection ranges, they could be compared directly only in the overlapping size range, namely: 10-50 microns diameter. In the LTWT where the populations were recorded simultaneously, there was some good agreement between not only the numbers but also the type of nuclei (mainly gas bubbles). However, the scattering technique indicated somewhat fewer nuclei in most cases. In addition, the scattering technique indicated fewer nuclei in the HSWT than had been previously measured by holography. Although there was some discrepancy between numbers of nuclei, both techniques demonstrated (directly from holography and by inference from the scattering results) that the nuclei were primarily solid particulates rather than gas bubbles.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):14-18. doi:10.1115/1.3240765.

The effect of polymer additives on underwater jet cavitation has been studied using a special camera. Solutions of the drag-reducing additives, polyacrylamide and poly(ethylene oxide), at concentrations of 25 ppm, decreased the cavitation inception index and greatly changed the appearance of the cavitation bubbles. Solutions of the non drag-reducing polymer, Carbopol, produced cavitation bubbles having the same appearance on pure water and did not change the inception index. In pure water, the cavitation appearance resembles ragged groups of small bubbles with the overall impression of sharpness and roughness, but in drag-reducing polymer solutions the bubbles are larger, rounded, and of completely different appearance.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):19-27. doi:10.1115/1.3240775.

Cavitation inception of a vortex is difficult to predict. This is due in a large part to a confusion in the type of cavitation occurring, i.e., vaporous versus nonvaporous cavitation. In addition, the vortex structure is poorly defined in many cases. These two problems are particularly important for the prediction of cavitation inception in a vortex created in the low momentum fluid near the inner wall of a rotor. The purpose of this paper is to present the results of a vortex cavitation investigation which are both experimental and theoretical. A vorticity flow analysis is developed and employed to assess the effect of vorticity on cavitation inception of a vortex. Previous investigations have shown that the minimum pressure coefficient of a vortex depends upon the vorticity associated with the vortex. Employing secondary vorticity equations, the vorticity is calculated in the blade passage. Changes in passage vorticity are used in a simple vortex model to predict trends in cavitation inception of a vortex. Theoretical results indicate that small changes in vorticity distribution near the inner wall of the rotor create rather large differences in the cavitation inception of the vortex. These small changes are primarily due to changes in the secondary vorticity. This secondary vorticity dominates the vortex structure. Comparisons are presented between the predicted and measured cavitation inception and good agreement is shown when the effects of gas on cavitation inception are reduced. Experimental data confirms that secondary vorticity dominates the vortex structure. In addition, experimental cavitation data are presented which show the dramatic influence of a gas on cavitation inception of a vortex.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):28-32. doi:10.1115/1.3240776.

Earlier desinent cavitation studies on a 1/8 caliber ogive by one of the authors (J. W. H.) showed a sudden change in the magnitude of the desinent cavitation number at a critical velocity. In the present work it is shown by means of oil-film flow visualization that below the critical velocity a long laminar separation bubble exists whereas above the critical velocity the laminar separation bubble is short. Thus the desinent cavitation characteristics of a 1/8 caliber ogive are governed by the nature of the viscous flow around the body.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):33-41. doi:10.1115/1.3240778.

This is the first part of a study of “third” two-dimensional problem of three-dimensional blade systems of hydraulic machines. Proposed herein is a method of obtaining and evaluating the three-dimensional effect on a system of hydraulic machine blades with arbitrary geometry. An analysis method indicating the velocity distributions on surfaces perpendicular to the mean axisymmetric stream surfaces has been formulated to help create a general understanding and awareness of the flow conditions in the runner passage. An application of generalized analytic functions for inviscid, incompressible flow has been made to find out the general solution on an auxiliary plane, transformed conformally from the physical plane. Integral equation systems for the tangential velocity and the velocity potential function have been deduced. Thus, the solution of the three-dimensional flow problem is supplied by two-dimensional computation methods.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):42-51. doi:10.1115/1.3240779.

This is the second part of a study of the “third” two-dimensional problem of three-dimensional blade systems of hydraulic machines. Part I described the formulation of the problem and the proposed method of solution to determine the velocity field on surfaces orthogonal to mean axisymmetric stream surfaces. Part II presents the numerical method of solving the integral equations; a few numerical examples for actual impellers/runners are also given. The results are presented in a series of figures and tables showing the distribution of the velocity component c2 along the blade profile on the surface q1 = const. The purpose of these numerical examples is to demonstrate the method and to help create a general understanding and awareness of the flow conditions existing in the runner passage.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):52-58. doi:10.1115/1.3240780.

A universal wake Strouhal number, St* =fs d′ /Ub , has been proposed and is based upon the Strouhal frequency fs of the incident flow, the measured wake width d′ at the end of the vortex formation region, and the mean velocity Ub at the edge of the separated boundary layer. This universal parameter collapses these characteristic parameters for bluff bodies onto a single curve for wake Reynolds numbers between Re* = 100 and 107 . The pressure drag, vortex shedding frequency and base pressure are related through an inverse dependence between St* and a wake drag coefficient CD * = CD /(d′ /d) K2 , where K = (1−Cpb )1/2 . The product St* CD * is equal to a constant value of 0.073 ± 0.005 for Re* in this same range of Reynolds numbers.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):59-66. doi:10.1115/1.3240781.

The overall objective of this experimental program was to quantify the effects of rotor-stator axial spacing on the fundamental time-variant aerodynamics relevant to forced response in turbomachinery. This was accomplished in a large-scale, low-speed, single-stage research compressor which permitted two rotor-stator axial spacing ratios representative of those found in advanced design compressors to be investigated. At each value of the axial spacing ratio, the aerodynamically induced fluctuating surface pressure distributions on the downstream vane row, with the primary source of excitation being the upstream rotor wakes, were measured over a wide range of compressor operating conditions. The velocity fluctuations created by the passage of the rotor blades were measured in the nonrotating coordinate system. Data obtained described the variation of the rotor wake with both loading and axial distance from the rotor as parameters. These data also served as a reference in the analysis of the resulting time-variant pressure signals on the vane surfaces.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):67-72. doi:10.1115/1.3240782.

The discharge coefficients for two inlet nozzles with different contraction shapes were studied experimentally together with the behavior of the boundary layer on the nozzle wall. For the inlet nozzle which has the same contraction shape as the ISA 1932 nozzle, it was found that boundary-layer transition with an intermediary separation bubble is responsible for the occurrence of a large hump in the discharge coefficients. The well-rounded inlet nozzle showed a satisfactory performance. The effect of pressure tap size and that of the proximity of a wall to the inlet on discharge coefficients were quantified experimentally.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):75-81. doi:10.1115/1.3240787.

A set of closed dynamical equations for the mean and variance of the turbulent vorticity field are used to study the piston driven flow in a two-dimensional model of an internal combustion engine cyclinder during the compression stroke. A dominant feature of the flow is observed to be the development of large corner vortices. In their vicinity intense production of turbulence occurs which locally greatly exceeds that resulting from the compression of intake generated turbulence. A comparison of the flow field at two strokes to bore ratios is made.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):82-87. doi:10.1115/1.3240788.

The effect of inertia on flow between a rotating and a stationary disk is studied for the case when the stationary disk is tilted with respect to the rotation axis. Based on the three-dimensional Navier-Stokes equations, an analytical solution is developed, with three perturbation parameters: aspect ratio, Reynolds number and tilt parameter. The noninertial hydrodynamic effect is to produce a moment which acts on the rotor and is perpendicular to both rotation and tilt axes. The effect of the inertia terms is to contribute another moment which acts on the rotor along the tilt axis. This analytical prediction explains qualitatively experimental and numerical results obtained at finite Reynolds numbers.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):88-95. doi:10.1115/1.3240789.

Model tests were performed in a wind tunnel to determine vortex shedding patterns induced around a circular cylinder by spanwise shear in transitional Reynolds number flow. In addition, mean and fluctuating pressure measurements were obtained. The introduction of shear in the upstream flow generated two distinct cells of vortex frequencies behind the cylinder in the transcritical regime, thereby documenting for the first time that the re-established high Reynolds number shedding closely parallels patterns already observed in subcritical flow. The two cell pattern did not permit any correlation between shear level and cell length to be found.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):97-103. doi:10.1115/1.3240793.

The center-line development of a turbulent boundary layer flow through a gap in an isolated wall mounted roughness element has been studied experimentally. The centerline flow downstream of the gap can be divided into a distortion region followed by a readjustment region. The nature of the distortion produced by the gap varies with gap size and thus flows downstream of large gaps were found to differ significantly from those through small gaps. After distortion the layer readjusts itself and approaches equilibrium conditions of an undisturbed zero pressure gradient layer. The readjustment starts near the wall with the turbulence adjustment preceding the mean flow adjustment. For flow through all six gap sizes the growth of the centerline internal layer can be described by a single function if internal layer height and distance from the gap are non-dimensionalized with the local wall length scale. Well downstream of the gap it is shown that all six centerline flows are similar and are approaching equilibrium conditions in a similar manner.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):104-111. doi:10.1115/1.3240755.

This paper gives a description of a calculation method for 3D turbulent and laminar boundary layers on nondevelopable surfaces. A simple eddy viscosity model is incorporated in the method. Special attention is given to the organization of the computations to circumvent as much as possible stepsize limitations. The method is also able to proceed the computation around separated flow regions. The method has been applied to the laminar boundary layer flow over a flat plate with attached cylinder, and to a turbulent boundary layer flow over an airplane wing.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):112-118. doi:10.1115/1.3240756.

It is important ot be able to calculate the lifetime of drops in central power station boilers so as to determine their penetration into the superheater sections of once-through boilers and as necessary input for the design of spray separators in recirculation boilers. The evaporation rate may be reduced by a large number of factors such as the elevation of the boiling point when the salt concentration in a drop becomes substantial or by the fact that the drop is evaporating in a medium with a density which may be more than a quarter of the water density. All the factors are considered and calculations suitable for deskwork are presented.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):119-126. doi:10.1115/1.3240757.

Measurements of four types of correlations in the self preserving region of a turbulent plane jet are presented. Negative correlations are noted between the two halves of the jet, exhibiting an apparent flapping like motion. That motion is noted to be self-preserving and most likely attributed to the presence of large scale organized structures.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):127-132. doi:10.1115/1.3240760.

The paper addresses the measurement and prediction of heat, mass, and momentum transport in a confined axisymmetric turbulent nonreacting flow of a jet in an opposing stream. The predictions are obtained by solving numerically the conservation equations of the mean flow and the transport equations of the kinetic energy of turbulence and its dissipation rate and the mean square temperature fluctuations. The predicted velocity field is in agreement with the experiment, but the predicted scalar fields point to the need of examining the employed model of a scalar turbulent diffusion.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):133-140. doi:10.1115/1.3240761.

Parameter identification by optimization forms the analytical basis for extracting road load information from well controlled coastdown tests. Specific attention is given to the effects of both biased and random errors as may be caused by environmental disturbances, road conditions, inaccurate parameteric input, or signal and instrument noise. As a consequence, experimental conditions can be identified and quantitative criteria established for either avoiding or correcting such errors. Test results are presented which support the methodology and conclusions.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):142-153. doi:10.1115/1.3240764.

Measurements of the velocity characteristics of a jet in crossflow are reported and encompass the entire mixing region. Laser-Doppler anemometry was used in the upstream region (x/D ≤ 6) where the turbulence intensities were larger and hot-wire anemometry in the downstream region. The scalar field, stemming from the injection of a trace gas in the jet flow, was also determined. Jet to cross-flow velocities of 2.3 and 1.15 were used. The results confirm and quantify the double-vortex characteristics of the downstream flow and demonstrate that this is associated with fluid emanating from the jet. The velocity maximum observed further from the wall than the double vortex is shown to correspond to freestream fluid accelerated by the pressure field. The mean-velocity profiles in the plane of the jet exit are shown to be far from uniform and the developing jet to be characterised by strong anisotropy associated with the acceleration of the freestream around the jet and into its wake. Probability distributions of velocity together with values of the Reynolds stresses, allow a detailed interpretation of the double vortex in the downstream region and indicate, for example, the larger magnitude of the crossstream fluctuations. The flow in the downstream region is also characterised by very different magnitudes of the shear stress and the non-coexistence of zero shear stress and zero gradients of mean velocity and turbulence energy.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):154-161. doi:10.1115/1.3240766.

This paper presents a theoretical and experimental study of two-dimensional, turbulent jet-boundary interaction. The jet is discharged parallel to and offset from a solid wall in the presence of a secondary freestream. An integral approach which makes use of the entrainment concept is used to predict the jet characteristics. The theoretical model departs from previous studies by relaxing the constant base pressure and radius of curvature restrictions and treating both as variables in the preattachment region. In general, good agreement is observed between experimental and theoretical predictions.

Topics: Turbulence , Pressure
Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):162-169. doi:10.1115/1.3240767.

The time averaged wake structure of three characteristic vehicle shapes viz. Estate, Fastback and Notchback was studied on the basis of flow visualization and wake surveys behind smooth quarter scale models in a wind tunnel. The models differed through their upper rear-end shape. Flow in the separation bubble at vehicle base and the subsequent formation of a pair of longitudinal vortices aft of this region is analysed. The kinetic energy of the rotational motion in the wake is evaluated to give a “vortex drag” rating for the vehicle shapes investigated. Effect of body details was assessed by a parallel set of experiments with detail models of same principal dimensions. Force measurements supplemented the investigations.

Commentary by Dr. Valentin Fuster
J. Fluids Eng. 1981;103(1):170-178. doi:10.1115/1.3240768.

Low mass, high speed trains may be in danger of being overturned by strong crosswinds. This paper examines the aerodynamic data required to estimate overturning wind speeds. The results of wind tunnel tests and a moving model experiment, including the effect of the turbulent wind, are described.

Commentary by Dr. Valentin Fuster

ERRATA

DISCUSSIONS

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