Recent theoretical and experimental studies on pipe flow problems of magnetic fluids in a magnetic field are reviewed. Pipe flow resistance increases largely with the application of a magnetic field. A theoretical model of particle aggregation is presented to explain the flow characteristics obtained in the experiments. As specific examples of pipe flows, gas-liquid two-phase flow of a magnetic fluid and oscillatory flow of a magnetic fluid plug are investigated.

View full abstract

An incompressible stationary flow past an arbitrary two-dimensional cylindrical body at low Reynolds number is analyzed by an asymptotic approach. First, the integral formulations are derived from the vorticity equation and the continuity equation. Second, simple integral equations with only one variable are derived from them. From these integral equations, asymptotic integral equations on the Stokes and Oseen regions are finally obtained by using the concept of both the significant integral operator given by Kida and the generalized expansions given by Skinner. Furthermore, using analytical results given by Kida, it is shown that no other local region exists.

View full abstract

An incompressible stationary flow past a two-dimensional cylinder at low Reynolds number is analyzed by an asymptotic approach. In the previous work of the present authors the non-linear integral expressions of the generalized Oseen and Stokes limits were derived, so the main purpose of the present paper is to show that from these expressions we can obtain the same aerodynamic results as earlier ones, examine the overlap hypothesis by applying Kida's extension lemma, and confirm Skinner's theoretical results on the drag force which has been noted in the previous paper. We further show that we can bridge the 1/log Re series expansions of the solution to the next order which is of order Re including 1/log Re series expansion. In the present paper, a simple expression of the drag force of the first approximation is derived and it is available only by determining the Stokes solution. From the results for a circular cylinder, we have the following : the overlap hypothesis is satisfied, the simple expression of the drag force agrees remarkably well with experiments in a wide range of low Reynolds numbers, and there is a missing term in the second-order approximation of Skinner's theoretical result.

View full abstract

An encapsulated liquid drop, which is made of an outer liquid shell and an inner gas, has received considerable attention because the encapsulated drop has many highly functional practical applications. However, the buoyancy-driven motion of the encapsulated drop at finite Reynolds numbers has not yet been clarified. In the present paper, the motion of a spherical encapsulated drop was studied numerically. The empirical equation of the drag coefficient for an encapsulated drop was proposed. Furthermore, the adaptability of the proposal equation for the drag coefficient of a bubble, a liquid drop and a rigid sphere at Re≤1000 was presented.

View full abstract

The entrance length Le of a laminar oscillatory flow in a square duct was defined as the maximum axial migration distance of fluid particles from the duct entrance. The entrance region was divided into three subregions, and the limits among them were determined according to the instantaneous axial position of a particle relevant to Le. Analytical solutions were presented for the cases where only a linearized axial equation of motion could reasonably be applied. For each subregion, the analytical solution was compared with the velocity profiles measured by means of a hot-wire anemometer. Close agreement of results was seen in each subregion.

View full abstract

The flow between concentric rotating cylinders, which is called the Taylor vortex flow, shows various types of flow. These flows are classified into three modes. One is called the primary mode, which appears when the inner cylinder is accelerated gradually, and the others are called the secondary normal and secondary anomalous modes, which appear with a sudden increase in the inner cylinder's rate of rotation. Unsteady three-dimensional numerical simulations considering the effects of the inner cylinder's acceleration have been performed in order to reveal the different processes that occur between the primary normal and secondary normal modes. The bifurcation causing these modes was observed under the same conditions as in the experiment. Transitions to wavy Taylor vortex flow from both modes appeared with increasing time. In addition, velocity and torque fluctuations for a certain point in the flow field and for the inner cylinder, respectively, are shown.

View full abstract

In the design of an LMFBR (liquid metal fast breeder reactor), it is important to know the movement of the free surface in a tank. Thus we developed a computer code for analyzing the two-dimensional turbulent flow with a free surface based on the VOF (volume fraction of fluid) and compared the computed results with those of the experiments. The measurements of surface location were made through digital processing of video images, and the velocity distributions were measured by the particle trace method and using the laser doppler velocimeter. The flow patterns and the locations of surfaces computed by this code agree with experiments.

View full abstract

Flow patterns and bubble-collapse pressures in a Karman-vortex cavity behind a circular cylinder are investigated for several Reynolds numbers and cavitation numbers. The time histories and intensity distributions of the pressures are respectively measured by using a pressure transducer and pressure-sensitive films. The cavitation aspects are also observed by high-speed photography. The bubbles frequently collapse at the instant when the Karman-vortex cavity separates from the cylinder or the cavity itself. High impulsive pressures with spike pulses over 60MPa occur in the separation or the vortex formation region. With a decrease in cavitation number σ, these regions are combined into one. With an increase in Reynolds number Re, however, the distributions of the impulsive pressures show the same tendency, because they have the same Strouhal number St, but the pulse count rate and magnitude of maximum local impulsive pressures increase.

View full abstract

Renormalization group analysis for, the turbulence problem is presented in this paper. This procedure does not require any experimentally adjustable parameters. Numerical values for relevant constants of turbulent flow can be derived : for instance, the Kolmogorov constant, Ck=1.605, for the inertial-range spectrum and the turbulent Prandtl number, Pt=0.7179, for high-Reynolds-number heat transfer. A differential turbulence (k-ε) model is derived in the high-Reynolds-number regions of flow with the algebraic relation ν=0.0846k²/ε. A differential large eddy simulation (LES) model is also derived, as is the Smagorinsky constant, Cs=0.00615.

View full abstract

Although it is generally known that the Karman vortex street consists of large-scale vortex structures with three-dimensional characteristics, the causes of the three-dimensional instability have not yet been elucidated. In this study, the three-dimensional characteristics of boundary layer separation around a circular cylinder were observed in water flow by using the laser light sheet technique in the range of Reynolds numbers from 170 to 1340 in a closed-circuit water channel and a towing water tank. As a result, it was determined that the three-dimensional instability of wake flow is based on the three-dimensional characteristics of boundary layer separation around a circular cylinder, which varies with the Reynolds number of the flow.

View full abstract

The equations for spectral tensor ψ_{2, ψ1}, ψ_{2, 3} and ψ_1.3 are derived, and the roles of terms in the equations are studied. Some terms can be explained by the same concept of spectral tensor for i=j, but others play different roles. Production terms dependent on Coriolis force and metric force occur, and are not due to mean flow tensor. In particular, the production terms appearing in the equations for ψ_{2, 3} and ψ_{1, 3} have opposite functions. The role of the pressure-velocity terms is the transport of spectrum components. Viscosity terms play a destructive role, and they are not dissipated into heat. Spectral tensor ψ_{2, 1} is closely related to ψ_{1, 1}, ψ_{2, 2} and ψ_{3, 3}. The relationship between ψ_{2, 3} and ψ_{1, 3} is also close due to production terms.

View full abstract

In our previous reports, it was clarified that fluctuations in the jet center have an important effect on opposed turbulent jet flow fields. In the jet center, where symmetrical flow fields are characterized, the turbulent structure will be different from those in other parts of the jet. In the turbulent opposed jet center, the fluctuating velocity plays a role equal in importance to that of mean velocity, static pressure and fluctuating velocity. The present paper deals with the influences of the variations of the fluctuating velocity, employing the grid in the center of turbulent opposed jets, to the impact position.

View full abstract

Using the moire-schlieren method for symmetric and asymmetric flow fields, density distributions of underexpanded free jets of air from circular and square nozzles have been measured. Experimental moire image data of the flows, which show integrated light deflection across the flow, were converted to an internal density distribution by the numerical inverting method. The experiment was performed systematically by changing the pressure ratio across the nozzles from 4 to 13. It was found that the density profiles measured in the expanded free jets from a circular nozzle agreed well with other experimental results, and that those for a square nozzle were much more complex than for a circular nozzle. In spite of the large difference in such profiles, the density variations along the flow axis for both nozzles were very similar to each other.

View full abstract

An experimental and numerical study was conducted on converging cylindrical shock waves. The goal of the present study was to clarify the nature of movement and instability in converging cylindrical shock waves. Experiments were conducted in an annular shock tube of 230mm outer diameter and 210mm inner diameter, connected to a cylindrical test section of 210mm diameter. Double-exposure holographic interferometry was used to visualize the converging cylindrical shock waves. Incident shock Mach numbers ranged from 1.1 to 2.0 in air. A numerical simulation was conducted using the total variation diminishing (TVD) finite difference scheme. It was found in the experiments that the shock wave configuration was initially cylindrical, but gradually deformed with propagation towards the center and finally showed mode-four instability. This is attributable to the existence of initial disturbances, which were introduced by the struts supporting the inner tube of the annular shock tube. This trend was significant for stronger shock waves, indicating that at the last stage of shock wave convergence each initial perturbation of the converging cylindrical shock wave was amplified to form the triple point of a Mach reflection. Numerical results correctly predicted the experimental trend.

View full abstract

This paper deals with effects of waveforms on predicting performance of a radial turbine for turbochargers under pulsating flows. The extensive unsteady tests were conducted with several pulse shapes and pressure amplitudes over a pulse frequency range from 10 to 70Hz produced by a disk-type pulse generator rig. Five kinds of pulse shapes were simulated for exhaust gas flows from diesel engines. The time-mean method and the quasi-steady flow analysis were carried out to predict the turbine characteristics by using the steady flow turbine performance. As the results, the following points were clarified. For most conditions, the quasi-steady flow analysis of turbine performance is valid even if the turbine is affected by the waveform, amplitude and frequency. It is considered that the turbine under pulsating flow conditions works along the curves of steady flow performance except for windage states. The timemean method predicts the power output fairly well, but grossly overestimates the mass flow rates for some waveforms.

View full abstract

Flow measurements were made in an unshrouded and a shrouded impeller with splitter blades by using a five-hole pressure probe. The passage flows and characteristics of these impellers were compared with those of impellers without splitter blades. In the pressure-sided passages of impellers with splitter blades, the wake positions are similar to those in impellers without splitter blades. In the suction-sided passages of impellers with splitter blades, the wakes of the unshrouded and shrouded impellers occur on the suction surfaces, but the relative velocity at the unshrouded impeller exit is smallest around the middle blade-to-blade position near the casing. In impellers with splitter blades, the blade loadings tend to become smaller, and the absolute circumferential velocities and total pressures become considerably larger than those in impellers without splitter blades. However, the effect of the splitter blade on static pressure differs between the unshrouded and shrouded impellers.

View full abstract

This paper is concerned with the practical application of self-tuning control on the position control of a pneumatic servo cylinder. In this study, instead of establishing a nonlinear mathematical model of a pneumatic servosystem, system identification by the recursive least square method is applied to find the mathematical model, then a self-tuning controller is designed and implemented in a microcomputer to control the position of the cylinder. A desired time response of the cylinder position can be obtained by using a self-tuning controller with the pole-placement approach. The experimental results with the self-tuning controller are compared to those from the conventional PID controller. With the self-tuning controller, the effect of the variable load and the variation of the system parameters are reduced.

View full abstract

In order to further develop appropriate general methodologies for analyzing the properties of reciprocating engine turbulence and to enhance a fundamental understanding of the in-cylinder turbulent flow phenomena, basic aspects in the correlation and frequency spectral structure of engine turbulence were considered, particularly the estimation of a characteristic time scale, the macro time scale, or simply the time scale, on the analogy of the integral time scale of turbulence for stationary flows. Applications to the study of turbulence properties in an automotive engine under variable swirl flow conditions are presented, including an insight into the probability density function of the engine turbulence. In addition, evaluation of cross terms, which arise in the momentum equation when unconventional averaging is used for the in-cylinder velocity, was provided with reference to the cycle-resolved data reduction procedure previously developed.

View full abstract

Measurements of mean values and turbulent intensities of velocity and temperature fluctuations of grid turbulence have been made in a developing region just after an electrically heated screen. The air velocity is set as 1m/s. The electric current to the screen is adjusted to be the downstream temperature (Ta) 32°C, 100°C or 150°C. Hot-wire anemometry with compensation for temperature variation is used to measure velocity with temperature variation. In a developing region, both velocity and temperature distributions become non-uniform at a point just downstream of the heated screen. The distributions recover gradually to the uniform condition in a downstream region. Turbulent energy dissipates with the distance from the heated screen. The dissipation rate is decreased with increasing air temperature.

View full abstract

As one type of particle beam processing, electron beam machining is quite popular for its high efficiency of energy conversion and its ease of beam production. In order to evaluate energy transfer by an electron beam qualitatively, the method of numerical simulation of high-energy electron behavior in substances is investigated. In this paper, a Monte Carlo simulation is presented in which high-energy electrons are decelerated in the course of atomic excitation involving Rutherford scattering by atomic nuclei. Beam energy here is transferred into the lattice system of a substance as a form of heat energy. Electron number density distribution in the substance as well as loss density disribution due to electron deceleration is calculated here. Although the shapes of these two distributions are similar, penetration depth of loss density proved to be about 30% smaller than that of electron number density, which was in fairly good agreement with the referenced data. Such a difference, which has never before been discussed in detail, is clarified numerically. In addition, beam reflection on the surface is of great interest, and numerical simulation, as shown in a previous experimental work, proved that power reflection is slightly less than mass reflection.

View full abstract

Pool boiling experiments under microgravity were conducted, utilizing parabolic flight maneuvers with a CNES Caravelle 6R aircraft. The experimental apparatus taken aboard the aircraft was constructed to enable simultaneous recording, by a video camera, of the side view of vapor bubbles generated on Joule-heated titanium foil immersed in a pool of n-pentane liquid, and the color pattern exhibited by a liquid-crystal layer, which was plated on the backside of the foil. This apparatus successfully recorded observations of behaviors of vapor bubbles, at various liquid subcoolings (7-32K) and heat fluxes (1.1-42kW/m²), and spatial color variation of the crystal layer, possibly reflecting microlayer dynamics beneath the individual bubbles. In spite of the great difference in size and population of the bubbles, between microgravity and terrestial conditions, we found no more than a moderate difference in overall heat transfer between the two conditions.

View full abstract

The colors of propane/air premixed flames were quantitatively characterized by the chromaticity coordinates (x, y) defined by the CIE 1931 standard colorimetric system. Detailed relations between flame colors and flame spectra were investigated in the range of air ratio from 0.7 to 1.2. The experimental results indicated that the chromaticity coordinates are useful to accurately express the change of flame color caused by the variation of air ratio. It was also found that the flame colors are mainly attributed to the relative spectral intensity of the CH bands, the C₂ bands and the continuous spectrum and that the continuous spectrum plays an important role in the color determination, especially for the lean mixture. Moreover, it was suggested that flame color can be used to evaluate the air ratio and the relative spectral intensity. Thus the flame color could possibily be a valuable information source for combustion diagnostics, and is applicable to combustion control and monitoring systems.

View full abstract

A binary gas permeation method is proposed to investigate gas transportation in the components of solid oxide fuel cells, in which the properties of porous media, such as pore diameter, porosity and the tortuosity factor, are obtained from the measured permeabilities of two different gases. It is also confirmed experimentally that permeabilities can be predicted with high accuracy for other gases as well as for nonisobaric binary gas diffusion systems. Moreover, we consider the airtightness of plasma-sprayed yttria-stabilized zirconia films and estimate the concentration polarization and distribution of total pressure in porous media at a high temperature.

View full abstract

The atomization characteristics of an atomizer of the ultrasonic vibration type are investigated when fuel is supplied intermittently to the vibration member. The time interval for the liquid supply is 100ms and the supply is varied from 5ms to 20ms. The mean diameter, velocity, and flow rate of the spray are measured by a phase doppler-type analyzer. The velocity is 0.2-1.0m/s and the mean diameter is 70-90μm. When the fuel is supplied intermittently to the vibratory member, large-size droplets are formed initially. The droplet size is reduced by improving the vibratory member shape and controlling its vibration.

View full abstract

Effects of flame stretch and flame curvature on the extinction of turbulent premixed flames in a stagnation point flow have been studied experimentally. A lean propane/air mixture was used. Bulk stretch rate was varied from 15s^-1 to 60s^-1, while the turbulence intensity of velocity fluctuation in the approach flow was varied up to 0.6m/s. Near the extinction limits, the local stretch rate was estimated by measuring the mean centerline velocity with LDV and the local flame curvature was measured using a laser tomographic method. The local stretch rate due to flow divergence decreases, while the stretch rate due to the local flame curvature increases with increase in the turbulence intensity. As a result, it is proposed that the sum of those two stretch rates plays a key role in flame extinction.

View full abstract

The instantaneous IMEP method is advantageous in that it is feasible for piston friction measurement with minor engine modification, while its weakness is the low accuracy achievable in the end result. Because of error accumulation from various input data measurements and ensuing data processing, its result is bound to have a low confidence level. For minimization of the error, the frequency response function was determined for each measuring system, and the relative time delays among the measured signals were obtained by the application of the cross-correlation function. In order to further increase the accuracy, an accelerometer was installed at the piston skirt for direct inertia force measurement of the piston assembly. In addition, the crank case pressure was measured. Relatively reliable friction data were obtained up to 2000rpm motoring and 1500rpm firing conditions after taking these appropriate measures.

View full abstract

From the viewpoint of energy conservation and the development of new energy resources, it is important to utilize geothermal resources and waste heat from factories. Among energy conversion device, there is a radial outflow reaction turbine, i.e., Hero's turbine. Performance characteristics of Hero's turbine are analytically and experimentally clarified for flashing expansion of initially subcooled hot water. It is found that: (a) there is an optimum number of revolutions at which maximum turbine efficiency can be obtained; (b) Hero's turbine internal efficiency can be expressed as an algebraic equation and (c) nozzle loss accounts for almost 90% of the total turbine loss.

View full abstract

The present study aims to develop an optimal wind turbine generator system with a storage battery for use in developing countries. In order to utilize wind energy efficiently, the wind turbine output should match the battery load. If we could control the revolution of the turbine in proportion to wind velocity, we could obtain the optimal system. However, the battery load suppresses increased revolution because the voltage is fixed. Therefore, we must compromise with an optimal fixed revolution. The present paper proposes a method to determine the voltage and the capacity of the battery based on computer simulation. We have applied it to our Darrieus-Savonius wind turbine (1kW) and have confirmed its validity with a 4-month field test.

View full abstract