This review describes methods and techniques developed to date for measuring elongational stresses produced in elongational flows. Experimental results are presented and it is shown that in many cases the strain exerted on a fluid element is important for the stress in elongational flow fields of dilute and nondilute polymer solutions. In several cases, however, it is demonstrated that the elongational stress is correlated with the elongational rate and agrees with the prediction of the FENE dumbbells model of the second-order type.

View full abstract

The cellular automaton method, which can generate the Navier-Stokes equation, is utilized for simulating fluid motion numerically. This method has the advantage of allowing examination of the statistical property involved in the macroscopic variables, since the elementary process is described as a stochastic process. The present paper demonstrates that an automaton solution of Poiseuille flow results in turbulent behavior with a Reynolds number of 3200. After careful review of the calculated intensity of turbulence and correlation function of this solution, it may be concluded that this method serves as a reasonable alternative to the Large-Eddy-Simulation (LES) and k-ε methods for examining turbulent flow.

View full abstract

This paper presents a finite difference scheme for solving incompressible viscous flow around an arbitrary oscillating body. This scheme is an extension of the SMAC method to a moving curvilinear coordinate grid. The scheme comprises two stages. At stage one, the new time step velocity is calculated in the old time step coordinates. At stage two, this calculated velocity is reobserved in terms of the new time step coordinates (rezoning). This scheme is applied to calculation of two-dimensional flow around an oscillating flat plate, which is a fundamental model of the fin movement of certain kinds of fish or cetaceans or the flutter of an airplane wing. The solution is compared with that from the discrete vortex method. In particular, the characteristics of thrust and propulsive efficiency of the oscillating flat plate as a propulsion mechanism show good agreement with those from the discrete vortex method.

View full abstract

Incompressible flows around a yawed circular cylinder at a Reynolds number of 2000 have been calculated by the finite difference method based on the third-order upwind scheme for approximating the nonlinear terms of Navier-Stokes equations. No turbulence model is employed in this study. Computations are carried out for cases with and without end plates in order to investigate the effect of end plates on the flow. The calculated pressure distributions around the cylinder are compared with experimental data and the agreement is satisfactory. The results of this computation show the details of the flow especially in the wake region and it is found that the existence of the end plate greatly affects the location of the separation point.

View full abstract

Chaotic mixing in two-dimensional square cavity flows with a periodic lid velocity was numerically simulated. To quantify the mixing in the cavity, the pseudo-Lyapunov exponent λ was measured. After selecting Reynolds number (Re) and amplitude of the lid velocity (A) as parameters which govern the flow field and mixing conditions, relationships between these parameters and λ were investigated. The obtained results showed that neither Re nor A affects the value of period T which gives the maximum λ.

View full abstract

This paper presents the result of Large Eddy Simulation (LES) of a spatially developing plane turbulent jet flow. Because simulation of the flow is sensitive to numerical disturbance due to inadequate treatment of outflow velocity boundary condition just as the flow is easily influenced by external disturbance in experiments, it prevented from the success for simulating the flow. Here a new outflow velocity boundary condition of LES is proposed based on physical insight of the spatially developing plane turbulent jet flow. The result of LES using this outflow boundary condition exhibits that many essential features of the unsteady spatially developing plane turbulent jet flow are reproduced by the present computation.

View full abstract

Until to the authors' previous reports on the oscillation of circular shock waves, investigations have been concerned with situations where the upstream flow is uniform, and oscillation and deformation are induced by only downstream conditions. However in the centrifugal compressor, the flow into the diffuser becomes nonuniform due to the impeller wake and the stall in the upstream impeller, which causes deformation and oscillation of the shock wave. Here, the above effects are considered. The upstream disturbance was generated by cylindrical bars. An imperfectly circular shock wave was induced by the wake, and the oscillation state, along with the modes caused by forced oscillation, was investigated experimentally. It was found that the basic mode of oscillation is predominant and that the oscillation is weaker than in the case of a uniform upstream.

View full abstract

In order to investigate the accuracy of finite difference methods for weak shock waves, comparisons for the reflection and the diffraction of weak shock waves around wedges and corners between simulations and analyses are made. Analytical solutions are obtained for the wave equations by employing the conical flow method of Keller and Blank. Finite difference simulations are conducted by applying such finite difference methods as the total variation diminishing (TVD) method, the flux-corrected transport (FCT) method, the flux vector splitting (FVS) method and the method of Roe to numerically solve the Euler equations. Among the finite difference schemes examined in the present investigation, the TVD method is the most reliable because of its stability and smooth resulting curves.

View full abstract

Shock wave propagation in polyurethane (PU(R)) foam was experimentally studied. The experiment was conducted in a shock tube by measuring pressure along the PU(R) foam in a shock tube, by means of holographic interferometry and streak camera recording. It was found that the stress-strain curve of PU(R) has an inflection point. When the pressure behind the incident shock wave was below the inflection-point pressure P_c, the wave impedance ratio of the incident shock wave and transmitted pressure wave Z_i/Z_t increased sharply. This indicates that only a small amount of energy is transmitted into the foam if the pressure is below P_c. For pressures larger than P_c, Z_i/Z_t decreases asymptotically to a small finite value.

View full abstract

When a high-speed train such as the Shinkansen enters a long tunnel, a pressure wave is generated at the nose of the train causing a sonic boom at the tunnel exit. This noise, which can be very disturbing to nearby residents, is strongly dependent on the speed of the train. Hence, a bottleneck has developed in the quest for increased rail capacity through the use of even faster trains. In order to reduce sonic boom, the authors have proposed passing the pressure or weak shock wave through a density gradient field perpendicular to the axis of a tunnel and located at its exit. In this report, the behavior of a weak shock wave passing through such density gradient fields is examined by means of both computer simulation and experiment.

View full abstract

The flow field of a supersonic flow CO chemical laser is simulated solving two-dimensional Navier-Stokes equations, and the effects of viscosity, shock waves and the mixing process on the small signal gain coefficients are studied. The calculated flow pattern agrees very well with the chemiluminescence from the laser cavity which was observed previously. Both the calculation and the experiment show that the O₂ streams and the dissociated CS₂ streams collide with each other at the nozzle exits, generating oblique shock waves. It is also shown that mixing of the O₂ streams and the dissociated CS₂ streams is significantly slow, and the temperature and gain distributions in the mixing layer are strongly influenced by the shock waves and expansion waves. The calculated distributions of the small signal gain coefficients along the flow agree fairly well with the measured values in the region near the nozzle exits where the three-dimensional effects are relatively weak.

View full abstract

The motion of a single bubble is simulated directly using the full conservation equations for mass, momentum and energy when the surrounding pressure increases stepwise. The numerical results for several cases reveal that considerable distributions of temperature and concentration of vapor are formed inside the bubble, and heat and mass transfer inside the bubble have a great influence on the bubble motion. Heat transfer on the bubble wall, which is mainly governed by the temperature gradient, causes the thermal damping of the bubble motion. This damping behavior is influenced by the initial bubble radius. Mass transfer on the bubble wall, which is influenced by diffusion phenomena between vapor and noncondensable gas, also greatly effects the bubble motion.

View full abstract

The present work investigates the dynamics of a cluster of bubbles in a liquid by means of a series expansion of spherical harmonics. The governing equations of three-dimensional motions for an arbitrary configuration of N bubbles are derived by taking account of the translational motion and the deformation of each bubble induced by mutual interactions among the bubbles. These equations are exact to the order of (R_IC/L_IJO)⁵ for inviscid terms, where R_IC is the characteristic radius of a specified bubble I and L_IJO the initial distance between the centers of the bubbles I and J. Viscous effects of the liquid are considered up to the first order for quantities of the translational motion and the deformation, on the basis of the potential solution. The equations involve previous results for a single and two bubbles as special cases. The characteristic equation for oscillations of N spherical bubbles is also obtained, and natural frequencies are calculated for specified configurations of the bubbles. It is shown that the lowest natural frequency of the bubbles is much lower than the frequency of an isolated bubble.

View full abstract

Some factors concerning the method of cavitation inception measurement are experimentally investigated. They are definition of cavitation inception, effect of air content, effect of methods of pressure change in cavitation testing and its rate of change. The test body is a circular cylinder in the flow range of subcritical Reynolds numbers, where cavitation bubbles occur in the separated shear layer behind the cylinder with a large scatter of incipient cavitation numbers. In the present work, one of the most effective definitions is proposed for cavitation inception and the following results are obtained using the proposed definition : (1) the inception values are greatly reduced in water with low air content of less than a critical value, (2) the inception point is affected by the pressure change rate under the condition of variable flow velocity. The uncertainty in cavitation inception measurement is also shown from a large number of measurements.

View full abstract

Depending on the preset conditions, a flashing liquid jet can either shatter or remain in the form of an irregular continuous column after being released from the nozzle to the flashing chamber. Four physical characteristics of a flashing liquid jet in the field of low pressure have been unveiled, namely, non shattering liquid jet, partially shattering liquid jet, completely shattering liquid jet but in stagewise sequence and flare flashing liquid jet. Temperature distribution of the liquid jet along the axial direction is determined to be exponential and varies according to the liquid jet type. Other analyzed characteristic trends include radial temperature distribution of the liquid jet, droplet size distribution and volumetric flowrate distribution in the dispersed jet. In addition, the contribution of bubble nucleation and growth to the liquid jet shattering in the field of low absolute pressure is hereby made clear.

View full abstract

Bingham plastic fluid flows through coiled pipes of circular cross section were experimentally studied. In the investigations, kaolin clay slurries were used as the Bingham plastic fluid. The effects of the Reynolds number Re or the Dean number De on the friction factor λ_c in the laminar steady flow of the Bingham plastic fluid flowing through coiled pipes were clarified. The apparent Reynolds number Re_ap, expected to play the equivalent role of Re in Newtonian fluids, was introduced using the similarity law of energy dissipation which was derived from a generalized equation of non-Newtonian fluid. It was clarified that the effects of the curvature of the pipes are greater in the case of greater concentration of the slurry, and that there exists a good correlation between the friction factor and the apparent Dean number De_ap(defined according to Re_ap).

View full abstract

In the present paper, using a new finite-element method for incompressible electrically conducting fluid, transient and steady entrance flows are simulated. The GSMAC method is employed as the main algorithm of calculation to satisfy the conservation laws of both mass and magnetic flux. Entrance flows between two parallel plates under a uniform magnetic field are calculated to verify this method. Meanwhile, the entrance length can approximately be expressed as an explicit function of the Hartmann number, using the Oseen approximation and the momentum integral method. Numerical and theoretical results obtained here agree well with other numerical results or exact solutions.

View full abstract

Valve noise due to cavitation has frequently posed serious problems with the increase of operating oil pressure. For different electroproportional relief and flow control valves, we have carried out noise tests and numerically simulated oil flows under the assumption of noncavitating conditions in various configurations of the valves by a finite difference method. In addition, we have performed pressure measurements at two locations in a valve chamber, and flow visualization, employing two-dimensional models. By comparing measured and computed results, some design criteria can be set up for low-noise valves.

View full abstract

The idea of tensile stress in a liquid is not necessarily familiar in fluid engineering, and much less accepted is the belief that a tensile stress can be propagated through a liquid. The present paper offers evidence of those `tensile waves' in a hydraulic oil. This experiment showed that a pipe with one end blocked, branching from the main line where column separation is started, creates transient waves exhibiting alternate tension and compression. Ordinary viscous wave theory proves to properly predict the measured pressure waves, which reveals that the propagation velocity of a tensile wave is no different from that of a compressive one. Visual observation with the use of a transparent acrylic tube as a branch has confirmed that an oil column ruptures under excessive tension at the blocked end, and after rupture a minute bubble probably born through diffusive gas release from the oil remains.

View full abstract

In this investigation, the authors studied the stability of steady flows in collapsible tubes of which static characteristics were theoretically calculated in the previous paper. The dynamics of the collapsible tube is approximately represented by a simplified lumped parameter model. On the basis of the model, the instability condition has been derived and the mechanism of the instability of the steady flow has been clarified. The instability region obtained theoretically is qualitatively in good agreement with the experimental results.

View full abstract

The flow on a flat-plate wing with low aspect ratio is affected by the side wall boundary layer and also the wing tip vortex. In this study, the flow on the suction surface of the wing is observed by means of the oil-film flow-visualization method, and the pressure distribution on the wing surface along the chord is measured at many spanwise positions. The results obtained are as follows. The side wall boundary layer induces secondary flow which suppresses the separation of the flow near the root of the wing, and the wing tip vortex suppresses flow separation near the tip of the wing. On a wing having a small aspect ratio, the separation of flow is suppressed more strongly by the effects of both the side wall boundary layer and the wing tip vortex.

View full abstract

This paper describes the rotation speed control of a horizontal-axis wind turbine. A tip vane has the excellent capability of improving the performance of the horizontal-axis wind turbine. Also, we found that the rotation speed of the wind turbine is controlled by changing the sweep angle of the tip vane. The relationships among the sweep angle of the tip vane, the change of wind speed and the rotor speed are investigated. As a result, a method to maintain constant rotor speed for changing wind speed, by means of an electromechanical apparatus, is developed.

View full abstract

The objective of the present study was to investigate the suppression of the fluid forces acting on a circular cylinder by controlling the flow around it. Flow control was established by introducing a fine flat plate in the neighborhood of the main cylinder. Measurements were carried out with variation of the position and attack angle of the flat plate at a Reynolds number of 6.5 × 10⁴. The time-averaged and fluctuating fluid forces were examined to estimate the magnitude of the reduction and to identify the optimum position of the control flat plate in reduction the fluid forces. Furthermore, the mechanism of the flow control, the nature of the controlled wake, the stability of the flow control and the relationship between the characteristics of the controlled fluid forces and behavior of the wake were discussed in detail. Then, the applicability of the present method reduction of the fluid forces and suppression the vortex shedding was evaluated

View full abstract

In an attempt to improve the aerodynamic performance of axial-flow turbines, boundary layer fences were attached to the blade suction surfaces and the endwalls of a rectilinear turbine cascade. Measurements of three-dimensional flow and total pressure losses were carried out at various protruding heights and locations of the combined fences. The best combination resulted in a considerable attenuation of secondary flow, an improvement on the spanwise distribution of outlet flow angles, and a 26 percent reduction in the gross loss of total pressure. This reduction of loss was larger than either the maximum reduction with the blade fences only or the maximum reduction with the endwall fences only. The best combination was distinguished from a straightforward combination of the optimum blade fences and the optimum endwall fences .

View full abstract

Basic experiments were carried out concerning the possibility of continuous ice making in supercooled water in a forced flow. In the experiments, the critical conditions of ice formation and stability of supercooled water flowing in cooling tubes were examined for four sets of tube dimensions under various Reynolds numbers. It was established that two ice growth modes of annular ice and dendritic ice appeared in the tubes according to the degree of supercooling. Nondimensional correlation equations of supercooling for ice nucleation in laminar and turbulent flow regions were derived as a function of the ratio of the thermal boundary layer thickness to the characteristic length. The latter is defined as the distance from the position where the mixing average temperature of the flowing water falls below freezing point, to the tube exit.

View full abstract

This article critically examines the widely accepted view that nucleate boiling heat-transfer data exhibit power law behavior, and that the ΔT exponent is approximately 3. The examination reveals that ● the power law view is based on induction methodology which is NOT rigorous ; ● the power law view is NOT supported by data in literature. Rigorous induction methodology reveals that nucleate boiling data generally exhibit LINEAR behavior, i. e., nucleate boiling data generally yield straight lines when drawn on a linear scale ; i. e. the ΔT exponent is generally unity.

View full abstract

The effects of free stream velocities, longitudinal pressure gradients and discrete wall jets on concave wall heat transfer characteristics have been examined in wind tunnel and water channel flows at momentum thickness Gortler numbers from 5 to 30 at the initial station. It was shown that longitudinal vortices became supressed with increasing free stream velocities. The pressure gradient parameters of up to 0.75 × 10^-6 resulted in a regular vortex structure and caused mean heat transfer coefficient to increase. The pressure gradient parameter of 1.8 × 10^-6 was sufficent to suppress the vortex development so that there was no heat transfer augmentation above that of flat plate values. Measurements with blowing parameters of up to unity showed that the streamwise vortices and the wall temperature patterns were controlled by the jets so that maximum Stanton numbers exceeded minimum values by a factor of around three. With blowing parameters larger than unity, the interaction between the stream-wise vortices and the wall jets caused irregularities in wall temperature patterns so that the streamwise variation of Stanton number remained nearly unchanged.

View full abstract

The soot formation during combustion is described by a theory based on cluster kinetics. Kinetic calculations in which carbon vapors (C₁ - C₅) and polyacetylenes are assumed to be the collision partners for cluster reactions support the formation by soot via soot clusters at higher temperatures. The prediction of the proposed theory agrees well with both the bell-shaped temperature dependence of soot yield and the rapid soot formation as observed from shock-tube experiments. It is also shown that the small clusters are mainly generated via C₂H, and they grow into large clusters by reaction with carbon vapors at moderately high temperatures, while they are decomposed into carbon vapors at very high temperatures. Furthermore, the effects of pressure and heterogeneous nuclei such as condensed PAHs on soot formation are discussed.

View full abstract

Experimental studies were conducted on the pollutant emission characteristics of the twin-flame matrix burner which is composed of a number of counterflow burners arranged in a regular matrix. The burner consists of two opposing burner matrices which form the combustion chamber with two side plates and a bottom plate. The profiles of temperature and concentrations of some stable species were measured at the exit of the combustion chamber for methane flames. The NO_x concentration in the combustion gas is far lower than the equilibrium concentration, because NO_x formation was suppressed in the vicinity of the combustion chamber exit. A burner of this type can be operated for the fuel-lean mixture and the fuel is burned completely because the counterflow twin flames are stabilized even in the mixtures near the lean limit of flammability. This means that the burner of this type with a large turn-down ratio is also useful for the suppression of pollutant emission.

View full abstract

Turbulence has a significant effect on the combustion process in an internal combustion engine. It is very important to separate the gas flow in the engine cylinder into turbulence and mean flow. Turbulence is defined as higher frequency components in the gas flow. In this study, some methods of frequency filtering to distinguish the turbulence components in the flow were investigated. The characteristics of four kinds of frequency filtering methods, that is to say, methods using both digital and analog filters of Butterworth types, and running average and stationary time-averaged methods, were compared. These methods were applied to an unsteady jet which Was similar to the flow in the engine cylinder, and the turbulence characteristics obtained were discussed.

View full abstract

For a typical reciprocating internal combustion engine with a piston, cylinder, connecting rod and crankshaft, there will be unnecessarily high side thrust acting on the piston in high-speed motion during the power stroke. This inherent side thrust will induce extra rubbing friction force and wear on the cylinder wall, thus reducing the engine power output. As a solution to this, the piston rings/crown can be innovatively redesigned to produce a transverse gas pressure to counteract the piston side thrust due to the tilting connecting rod. Thus the piston will move in the cylinder in a gas-cushioned manner to reduce adverse mechanical effects. Naturally, due to various engine limitations, this technology of gas-cushioned piston can be designed to be most effective only under certain operating conditions, such as a specific economical cruise speed of a ship or aircraft. In conjunction, the associated dynamics analysis and design of engine reciprocating mechanism is greatly enhanced by employing an ingenious varied-mass lumping method combined with the percussion concept developed by the author.

View full abstract