In order to prevent fatal cavitation troubles downstream from a cavitating branch duct and a drastic change in flow pattern, we carefully investigated the size distributions of bubbles flowing down from a right-angled branch duct for several types of observable cavitation, for various cavitation numbers and flow ratios. The distributions are very susceptible to the type of cavitation, but almost obey a long-normal distribution. These bubbles induce cavitation inception, and sometimes result in a drastic change in cavitation type from one to another, as a huge number of bubbles flow down even in the far downstream region.

View full abstract

In order to clarify the Johnsson effects which so often arise in single-foil cavitation-experiments, careful cavitation observations were made around a typical Clark Y foil of 4 percent thickness in our closed type cavitation tunnel, with which a number of cavitation-performance-tests have been done by late Prof. F. Numachi, for various cavitation numbers and incidences. For positive incidences of technical interest, we mainly observed spot cavitation, which seemed to be very insensitive to the cavitation nuclei, as well as flow fluctuations, accompanying the vortex type-, the band type-, the large spherical-bubbles as well as the bubble cloud. Abrupt separations on large groups of spot cavitation bubbles were also periodically observed within the transient region.

View full abstract

The growth of bubble in shear flow and uniform flow is studied numerically (with F. E. M.) and experimentally. The calculated results indicate that the growth rate of bubble in shear flow is not influenced by the condition of flow, and in the case where the bubble viscosity ratio is approximately one, the growth rate of bubble is equal to that of rigid gas-liquid interface. The experimental results show that the growth rate of bubble in shear flow is lower than that in uniform flow; and in shear flow and uniform flow the interfacial resistance against gas diffusion increases with decreasing the interfacial area of bubble.

View full abstract

This study investigates the growing process of an axisymmetric (bubble type) vortex break-down. In the preceding reports, the growing process of the bubble was discussed, based on the inviscid linear theory of the internal wave, and also examined by the numerical analysis of the Navier-Stokes equations. In this paper, experiments of the unsteady swirling pipe flow is conducted to confirm the theoretical results concerning the growing process. Generation of a group of internal wave by the abrupt change of the upstream boundary condition, its propagation and dispersion are measured by a Laser-Doppler-Velocimeter, and obtained data are compared with the results of inviscid theory. By the method of intermittent or periodic injection of dye, the growing process of a bubble is clearly visualized. The obtained results of the growing process showed a good agreement with the theory.

View full abstract

For designing a natural circulation boiler, a simple method for judging flow stability is reported in this work. The flow condition in a steady state is considered, and the relation between parameters which were already reported is made clear. When inlet throttling or outlet throttling is not provided in a channel, the number of parameters which decide the flow stability becomes two by the relation. Y_e, Which is one of them, is used for judging ; When the value "Y_e" is larger than about 11, the flow is unstable. When inlet throttling or outlet throttling is provided, the number of parameters becomes three, and so the stability maps are illustrated in the three dimensional coordinate.

View full abstract

Length scales were experimentally determined for an artificially induced turbulence field with large turbulent intensities. The longitudinal and lateral integral scales were much larger than the values for conventional grid turbulence; and Taylor's micro scale was almost the same. The streamwise growth-rates of the micro scales were in good agreement with the theoretical value for isotropic turbulence, but the growth-rates of the integral scales were smaller than the theoretical values. The degree of the axisymmetric anisotropy was evaluated on the basis of the Rotta's model. The process of the return to isotropy was gradually carried out towards downstream, and the values of the main diagonal components of the anisotropy tensor became almost unvaried for X/M > 60. The skewness factor was proved to be dependent upon the turbulent intensity.

View full abstract

Hot-wire measurements are reported for an initial mixing region of a plane turbulent jet. The investigation includes the measurements of the terms in the turbulent energy equation and the Reynolds transport equation. It is shown in the initial mixing region that the lateral distribution of all terms corresponds quite closely to the behaviour of the two-dimensional free shear layer. As the effect of the potential core disappears, the distribution approaches that for the established plane jet. However, it takes a considerable distance for the convection term in the turbulent energy equation to lose the characteristics of the free shear layer.

View full abstract

The authors reported previously that as the turbulence level upstream of the step became higher, the reattachment length became shorter. The objective of the present paper is to find out which turbulence in the transverse distribution at separation influences the reattachment process. Turbulence was generated by a grid at the nozzle exit, rod, or cavity near the step. Reynolds number based on step height was 3.2×10⁴. Maximum turbulence intensity near the wall at separation determines the reattachment length, and turbulence plays an important role in the entrainment region downstream of the step. Behavior such as a "turbulence barrier" is also observed at the edge of the shear layer.

View full abstract

This paper presents the results of an experimental study on the deflection and reattachment of a radial turbulent wall jet through two comparatively wide-spaced parallel disks, in the presence of a lateral control flow. The main content of this report concerns the effects of various opposite wall heights and control flow rates on the flow properties at and after the reattachment point on the wall surface. A flow field was divided into five regions. The changes of the maximum jet center velocity, the jet half-width and the turbulence at the velocity maxima, and also the deflection of the jet center axis in each region, depended mainly on the combination of the opposite wall height and the control flow rate. Further, the limiting position in which the wall jet flow pattern might be maintained was examined.

View full abstract

Swirling flow close to a wall has been studied. The measured velocity distribution near the wall indicated that the flow is hot skewed in this region. A mixing length model can be applied to this region of flow. When the swirl component is weak, the centrifugal force effect on the flow is small enough to neglect, and the flow is similar to that on a flat plate. As the swirl intensity increases, the centrifugal force has a noticiable effect on the flow structure, and the Monin-Obuknov formula is successfully used to calculate the flow.

View full abstract

A pseudo-unsteady system of equations for two-dimensional steady compressible potential flows has been solved by using the rational Runge-Kutta time-stepping procedure combined with the usual central finite difference approximations. Computed results for transonic flows past the NACA 0012 and RAE 2822 airfoils, and subcritical flow around a circular cylinder are compared with the Euler solutions. The potential solution is in close agreement with the Euler one for the case with weak shock waves. For stranger shock waves, significant discrepancy is recognized in the computed flow field due to the entropy production in the Euler solutions.

View full abstract

An analysis of the one-dimensional two-phase flow of magnetic fluid in a diverging duct is developed for a low-quality case, taking into account the slip and pulsation of gas bubbles. It is shown that the injection of gas bubbles in the throat effectively increases the pressure rise in the diverging duct under a nonuniform magnetic field. Furthermore, the analytical results are confirmed by an experimental study in which the effect of magnetic field on the total pressure loss in the test loop is measured.

View full abstract

To clarify the phenomena of particle aggregation in magnetic fluids, an analysis based on the tunnel theory has been developed, taking into account the effect of the magnetic interaction between clusters. It is shown that a stable colloidal dispersion as magnetic fluids is not achieved in the case of a strong magnetic interaction between particles, and that a phase transition occurs in increasing intensity of a magnetic field when the coupling coefficient λ is more than a certain value. Furthermore, it is found that there is no aggregation for water-based magnetic fluids even though the interaction between clusters is considered.

View full abstract

In the turbulent wake behind an upstream cylinder in a uniform flow, the downstream cylinder was set in the form of an intersection with the upstream cylinder. A secondary flow was found behind both cylinders. The important feature was found to be an increase of base pressure due to the formation of secondary flow patterns. The flow pattern has been verified by oil-film and tufts flow visualization. The base pressure at various spanwise stations for both cylinders was measured. Then, the relationships between the secondary flow and the base pressure was studied.

View full abstract

To apply the theory of meniscus instability to polymers and glassy metals, theoretical and experimental analyses of the flow with free surface were conducted. In this study, the relation between the modified capillary number and the minimum wavelength of disturbance was derived by the minimum principle of viscous dissipation energy. The relation was compared with the experimental results which were obtained by using glycerine and silicone oils of various viscosities. Agreement of the theoretical relation with the same experimental one was very good. The energy requirements for moving plates with respect to each other were experimentally given under various experimental conditions. The energy could be arranged by using the modified capillary number. Maximum load during plate movement was also investigated. Finally, the equation for calculating the energy from the minimum wavelength could be obtained in this study.

View full abstract

It is very interesting to examine fluid forces on an oscillating bluff body with splitter plates in stationary fluid in connection with wave-induced forces on ocean structures and others. The discrete vortex model together with the conformal mapping method has been a;;lied to analyze the flow around an oscillating flat plate with splitter plates in a stationary fluid. Numerical calculations are performed for different splitter lengths and inclined angles of plates in various oscillating conditions, and vortex patterns and unsteady forces are obtained. From the calculations, it has been found that the splitter prevents the development of the shedding vortices and that the forces on the plate are influenced by the splitter length and inclined angle remarkably.

View full abstract

The discrete vortex model with conformal mapping method has been applied to analyze the flow around the flat plate acting in rotational oscillation about the front end of a plate in a uniform flow. Numerical conditions are determined by comparing with the result of flow visualization already reported.. Then, numerical calculations are performed for various nondimensional frequencies (ratio of circumferential velocity at end of plate to uniform flow velocity) of more than 0.1 and amplitudes. Vortex patterns and unsteady forces are obtained. As a result, a thrust is generated on the plate for nondimensional frequencies of more than 0.1 by this oscillation. The thrust on the plate increases as the amplitude becomes large under the condition of these frequencies.

View full abstract

Sedimentation of small particles through liquid has been widely utilized to measure easily the size of the particles. Conserning this, experiments were carried out both for the case of a single particle falling through liquids and for the case where particles constitute a dispersion system and fall through liquids, and the following are clarified : (1) Drag coefficients measured on a single particle are approximately expressed with Stokes' law for creeping flow motion. (2) With sedimentation of dispersed particles, the falling velocity is in some cases abnormally low, though the concentration of the dispersion system used is hitherto thought to be too low to induce interaction among the suspended particles. (3) The abnomally low sedimentation speed is possibly due to the interaction among the suspended particles which are surrounded with electric double layers. (4) To suppress the occurrence of the abnomally low sedimentation speed and no measure the nearly correct size of particles, it is effective to add NaCl of the order of 1 mol/1 to the water of dispersion medium.

View full abstract

In order to improve jet-pump-performance, in this paper, the turbulent flow through a jet-pump-throat was carefully analyzed by using a simple flow model, and the results were compared with experiments on jet pumps from 8.3 to 100 mm in diameter d_t. The optimum throat length, l_23opt/d_t, was simply shown by l_23opt/d_t = A log Re + B where A, and B are constants, and Re is Reynolds number. We also found a superior pump with 40 percent of the maximum efficency.

View full abstract

We propose a new electro-pneumatic on-off valve with a disk flapper (a step-disk combining two semicircular disks with different diameters) driven by a pulse motor. This on-off valve consists of a disk flapper and a nozzle. The PWM mode is the technique used for driving the pulse motor. At first, as a preliminary experiment to establish the characteristics of the disk flapper, an eccentric disk is set to the nozzle exit and it is rotated by a motor. The effects of the position, the width and the diameter of the disk and the eccentric length to the back pressure are examined. Then, it is shown that pneumatic pulses to drive an actuator (air cylinder) are generated by continuous clockwise and counterclockwise rotations of the pulse motor, and the characteristics are measured.

View full abstract

The difficulty of making wave energy practical is to reduce the energy cost to an acceptable level. We have been studying the Pendulor type wave energy converter for this purpose, because it has excellent merits ; it is simple, highly efficient, and safe. We previously reported on a coastal experiment on the Pendulor installed outside the Muroran port. During the four years of the experiment, the Pendulor was often attacked by storms and was once broken after 20 months of operation. We investigated the cause of the damage and found the best way to prevent harmful power against the Pendulor. The Pendulor was improved by applying the measures derived from our investigations, which have made the wave energy system stronger and more efficient.

View full abstract

The effects of the size of a heated slab have been examined numerically on the characteristics of flow situation and impingement heat transfer caused by a laminar slot jet. Numerical values were obtained for the dimensionless width of a heated slab W = 0.5∼5.0, the dimensionless distance between the nozzle and the heated surface H = 0.5∼6.0 and the Reynolds number Re = 100∼1000. It was found that the generation of recirculating flows near the side surface of a heated slab and the upper confined plate depends on the ratio of the size of a heated lab and the nozzle width (W≷1) and the spanwise variation of the local Nusselt number on the lower wall exhibits one or two maxima.

View full abstract

An experimental investigation was made on heat transfer and pressure drop using a spacer in a rectangular channel, and the basic characteristics of the spacer were obtained. These included the local heat transfer coefficient and static pressure drop distribution. A heat transfer enhancement was caused by the effect of turbulence at the spacer outlet region, not as the effect of increased velocity in the spacer ; and then the flow diffusion and contraction around the spacer caused a larger pressure drop. Four dimensionless parameters were defined as spacer specifications : blockage ratio ε, gap ratio α, spacer length ratio β, and spacer existence ratioθ. Among the parameters, blockage ratio ε had the most influence on performance enhancement. The total performance between heat transfer and pressure drop were estimated under the condition of a constant fluid transportation power. The spacers with blockage ratio ε=0.5 and spacer existence ratio θ=0.2 gave excellent performances.

View full abstract

The effects of buoyancy on upward-flow laminar convection in the entrance region of inclined parallel plates are studied by a numerical method. Solutions are given for three thermal conditions of parallel plates : lower wall heated and upper wall insulated ; upper wall heated and lower wall insulated ; and both walls heated in the same uniform states. Results are presented to show how the developing upwasd flow and thermal field in the entry region are affected by buoyancy for various inclination angles of parallel plates. Correlation equations applicable in practice are also developed for the friction factor and the local Nusselt number.

View full abstract

Experiments on solidification blockage of 90° bend laminar flow were made with two R/a ratio double cylinders. The inner pipe of the cylinder was supplied a developed laminar water flow, and through the outer pipe flowed a low temparature coolant supplied from a reservoir. With the ordinary dimensionless axial coordinate for the straight pipe, two critical curves for blokage were obtained according to the R/a ratios. Available calculated and experimental results were used to see a proper dimensionless axial coordinate which could compactly represent curved pipe flow heat transfer, and ζ^* = (z/R)/Pr^m(De·a_R)ⁿ was selected with m=1 and n=1/2. The two critical curves fall into a single critical curve with the thus derived ζ^*₀ = (z₀/R)/Pr(De·a_R)^1/2 [u₀ = (π/2)·R for 90° bend].

View full abstract

For tube bundles of flooded evaporators, boiling heat transfer coefficients are improved by upward streaming vapor bubbles. The authors tried to clarify the relations between boiling heat transfer characteristics of single tubes and those of tube bundles. In this report, the effects of void fraction and bubble-rising velocity on boiling heat transfer characteristics are investigated. Experimental results are summarized as follows : (1) though bubble-rising velocity depends only on the total heat flux of lower tubes with void fractions below 90%, void fraction depends on both the total heat flux and the level of liquid when unheated ; (2) the effects of tube bundles on boiling heat transfer should be evaluated both by bubble-rising velocity and by void fraction ; (3) and effects of void fraction become larger under smaller temperature differences.

View full abstract

Field investigations on spectral hemispherical reflectances of typical natural and artificial surfaces in urban and rural areas (environmental surfaces) have been performed for solar irradiation in the wavelength range from 0.4 to 1.7 μm. Reflectance measurement is carried out accurately by a newly designed compact field spectrometer using an optical fiber and an integrating sphere with a wide view angle, regardless of the angular distribution of incident and reflected solar radiation. The reflectances of all environmental surfaces except snow cover are low in the visible region. A large difference is found between real surfaces and clean specimens. As the solar height decreases, the reflectances of vegetative fields become clearly higher in the infrared region. As the water content increases, the reflectances of permeable environmental surfaces become lower mainly in the infrared region.

View full abstract

For the thermal energy storage in an aquifer, the investigation of the structure of stratum at a moderate price is very difficult. Hence a field experiment is necessary to understand the general trend. A study of the thermal energy storage in an aquifer has been carried out on our site for 5 cycle. The experimental results showed the following. I) In the 5th cycle, the thermal recovery factor is expected to be 60%. II) IN the 5th cycle, the structure of the charge well was improved, reducing the heat losses from the casing pipe. The heat losses during charge were reduced to 2/5 of the previous one. III) The coefficient of performance of our solar system, based on the thermal storage utilizing aquifer, was 10.1. IV) The payback of the investment is 10 years when the recovery factor is 60% and the pump efficiency is 60%.

View full abstract

A numerical simulation model has been developed to predict the direct-injection stratified charge combustion in a constant-volume vessel. Important factors such as local fuel concentrations, their fluctuations and turbulent flow characteristics were measured throughout the vessel as functions of time. These data were utilized to estimate the burning rate composed of the turbulent fuel-air mixing rate and chemical reaction rate. The model can predict the combustion pressures and heat release rates measured for different ignition timings and spark locations.

View full abstract

Recently, two intake valve engines have come into wide use because of their high output performance. However, there exists some difficulties in strengthening the air motion in combustion chamber in order to achieve stable combustion on partial load in a lean mixture operation. In this study improvements in engine performance that result in a stable combustion of lean mixture are examined. These improvments are brought about by the installation of a swirl control valve in the siamesed-type intake port of a two intake valve engine. Characteristics of air motion inside a combustion chamber are also examined using a laser doppler velocimeter. As a result of this study it is clarified that lean stable limit was not always extended in proportion to the strength of tangential swirl and combustion velocity. It is also found that an enlargement of lean stable limit was related to a decrease of cycle by cycle variation.

View full abstract

All previous research efforts on the general purpose two-stroke cycle gasoline engine have been entirely concentrated on finding an effective way to inclease the charge of mixture into the cylinders. Unlike such traditional studies, our last report was focused on the method of separating exhaust gases from this type of engine. It is part of our ongoing research activities which deal with not only how to increase the charge of mixture into the cylinders but also how to make good use of exhaust gases by separating them into burned gases and a useful mixture. The present report, unlike the last one, does not focus on the separation technique but on detailed data on the opration of a four-stroke engine using the trial manufacture of the engine. The data include, among others, the relation-ship between the trial manufacture of the engine and the general purpose engine in performance, specific fuel consumption and emission gases (CO, HC, NO).

View full abstract

Diesel spray in a high-temperature and high-pressure atmosphere was recorded by pulsed-laser holography. The schlieren and shadow images of the identical spray obtained from a reconstructed hologram were analyzed. The concentrations of the fuel vapor and droplets were estimated by using the photographic densities of these images. In the radial distribution of the fuel vapor, the concentration shows an anticlastic pattern and is maximal between the center and edge of the spray. As the time advances from the start of injection, the fuel vapor concentration begins to show an angle pattern, and is maximal at the center of the spray. In the axial distribution of the fuel vapor, the concentration averaged in the cross section of the spray is maximal at the axial location a little upstream from the tip of the spray.

View full abstract

In the present studies, a new accumulation-type fuel injection system was developed to control the injection pressure independently of engine speed. The injection characteristics and the effects of the system on combustion characteristics represented by exhaust smoke density, specific fuel consumption and rate of heat release, in a small-sized direct-injection type diesel engine were investigated and the following was made clear : (1) Making use of the new injection system, exhaust smoke density was fairly decreased even at low engine speeds. (2) It was effective on the improvement of combustion characteristics to use the injection nozzle with smaller hole diameter corresponding to higher injection pressure, because of the increase in excess air ratio of fuel spray. (3) Increase of injection pressure made combustion characteristics insensitive to swirl intensity and improved specific fuel consumption, especially at retarded injection timing. (4)The shape of injection rate versus crank angle is important to reduce exhaust smoke density without a serious increase in maximum combustion pressure.

View full abstract

The effect of wall curvature on the full-coverage film cooling performance has been studied experimentally by using concave, flat and convex walls. The effect of coolant mass flux has also been investigated. A preliminary consideration, based on the balance of the centrifugal force associated with an injected fluid element and the pressure gradient normal to the wall, leads to the understanding that the trajectory of the injected fluid tends to move away from the concave wall and attach to the convex wall, respectively, under the present experimental conditions. This has been experimentally confirmed by both the boundary layer measurements and the smoke-wire flow-visualization. In accordance with the flow structure observed, the cooling effectiveness measured on the convex and concave walls is higher and lower than that on the flat wall. In addition, the mass flux ratio has little effect on the cooling effectiveness on the concave wall, while it is optimum in the range of 0.4 - 0.5 on the convex wall.

View full abstract