During the last decade, experimental and theoretical studies on fluid dynamics in diesel engines have seen considerable progress due to the development of laser techniques and the increase in computing power. However, there are some important problems which remain unsolved. This paper overviews recent works on in-cylinder air motion and the spray dynamics in diesel engined, and also summarizes the directions for future research. This review pays the attention to the current status of experimental techniques and the evaluation of multidimensional modeling. Necessary research in the near future concluded from the surveys includes more extensive assessment of multidimensional modeling and more benchmark experimental data for the assessment.

View full abstract

Transition in dilute polymer solutions was investigated in the same manner as outlined in the previous report for Newtonian fluids. First, we investigated the natural transition arising from a minimum critical Reynolds number. The results suggested that the structural alteration of a turbulent spot resulted in polymer drag reduction. Next, we maintained laminar flow above the minimum critical Reynolds number and visualized the growth of spots artificially created in the laminar flow. The results showed that polymer additives reduce the spreading angle of a spot. Striped streak patterns of ink were observed around the spots. This phenomenon is supposedly caused by the instability of polymer solutions and instability in the background flow.

View full abstract

An analysis is first presented to clarify the effects of a magnetic fluid film coated on a current-carrying cylinder on the free-stream structure and on the hydrodynamic drag force. The calculation is made for the case of Reynolds number Re=10 and a ratio between the mean film thickness and cylinder radius, ε=0.1, using a perturbation method taking ε as a parameter. Then, the effects of Reynolds number and film thickness on the film profile are experimentally studied. It is clearly shown that the drag of a cylinder is reduced by the presence of a magnetic fluid coating only when the ratio of the apparent viscosity of magnetic fluid to the viscosity of the outer fluid is less than 1/4.

View full abstract

This paper reports experimental studies on the dynamic behavior of a liquid in a cylindrical container subject to high-frequency vertical oscillations. The pressure response, surface wave motion and container wall response induced by vertical vibration were investigated. Measurements were made of the amplitude of the time-varying pressure, free surface oscillation and container wall vibration. It was found that the amplitude of pressure increased exponentially with time when the excitation acceleration exceeded a threshold value. The amplitude of pressure had a maximum at time t_c, which decreased as the excitation acceleration increased. It was observed that the variation of the amplitude of free surface motion and the container wall vibration with time showed similar trends as the pressure variation. These facts can be explained by the increase in the void fraction ; the wave velocity of the liquid-bubble-shell system approaches the resonant velocity as the void fraction increases. The time t_c was analyzed by treating the pressure wave as a one-dimensional acoustic wave. The experimental results of t_c were compared with numerical values, and good agreement was found.

View full abstract

The movements of small particles distributed uniformly in an unsteady flow in a square chamber having an inlet and an outlet are simulated numerically. The particles are classified into three types according to their final state : the exhaust type in which the particles are exhausted through the outlet ; the circulation type in which the particles circulate in the chamber, and the adhesion type in which the particles adhere to the walls of the chamber. The influences of the initial position of a particle, the Reynolds number R and the dimension ratio N characterizing whether a particle is coarse or fine, on the ratios of the number of particles in each type to the total number of particles are examined. The number of particles of the exhaust type has a maximum at a certain orbit. The particles of the adhesion type adhere to the walls in vicinity of the stagnation points.

View full abstract

Flow in a narrow gap along an enclosed rotating disk superimposed with through-flow is studied theoretically and experimentally. When the axial gap is narrow, or a large outward through-flow is imposed, the boundary layers on the rotating and the stationary walls interfere with each other. The present study proposes an analytical model for such interference of gap flow and gives a theoretical analysis which is easily applicable to various boundary conditions. For non-interference of gap flow, a theory giving better results than conventional theories is also presented and the critical gap between these two flow patterns is determined theoretically. It is shown that the radial pressure distribution is dependent mainly upon wall shear stress and radial expansion of the sectional area in the case of gap interference, and on the centrifugal force of the core flow in the case of gap non-interference. The coefficient of the disk friction moment is shown to be proportional to Re^-1/4 in the case of gap interference, and to be Re^-1/5 in the case of gap non-interference.

View full abstract

For radially-curved, mixed-flow, vaneless diffusers, the authors previously clarified that diffusers which have a small static pressure gradient normal to the walls produce a favorable inner flow and high performance. Using this concept in this study, a design method was established to produce a small static pressure gradient normal to the walls at any section in a radially-curved, mixed-flow, vaneless diffuser flow passage. Flow measurements showed that a favorable inner flow and high performance could be obtained using this method.

View full abstract

In the preceding paper, a two-layer flow model was proposed for analyzing the influence of tip clearance on the turning angle and the pressure loss of turbine nozzles, and the prediction was compared with a series of experiments at the rated nozzle angle. In the present experiment, the nozzle angle was changed so that the throat area was 0.8 and 1.4 times the road condition and similar experiments were repeated. Contour maps of total pressure loss and spanwise distributions of the mean exit-flow angle are presented. The theory based on the two-layer flow model predicts the effects of tip clearance well, although based on the two-layer flow model predicts the effects of tip clearance well, although it slightly underestimates the clearance effect for a lightly loaded condition. The reason for and a method to correct the underestimation are discussed.

View full abstract

the wear of various sleeve-type submersible bearings in slurry which are commonly used for diffuser-type vertical pumps was investigated. The test bearing materials were ceramics, tungsten carbides, and self-fluxing alloys which were expected to have sufficient wear resistance, even in slurry. Rubber, which is the most conventional bearing material, and resin were also tested. The test equipment simulated actual vertical pump, that is, test bearings were installed in slurry and the rotating shaft precessed. The slurry density was varied between 50 ppm and 3 000 ppm. It was found that wear rate and slurry density were strongly related in the range of slurry density from 200 ppm 3 0000 ppm. The materials of higher hardness had better wear resistance, in particular, the combination of silicon carbide and tungsten carbide showed superior wear resistance.

View full abstract

The melting process of a snow layer has been investigated for the case where heat is supplied from the bottom and the meltwater drains continuously. The melting was divided into three patterns, depending on the existence of an ice lens formed by the refreezing of permeating water. A one-dimensional melting model including the draining of meltwater was formulated to account for water permeation and its refreezing in a snow layer. The calculated results for the time variation of each layer and the rate of water drainage are in agreement with the experimental results obtained using compacted and granulated snow. It is found that the melting time and the melting efficiency are affected by water permeation and its refreezing, and the results calculated assuming the classical Stefan problem by neglecting water permeation predict a higher melting efficiency.

View full abstract

A theoretical investigation is presented for free convection in vertical concentric annular ducts with uniform temperatures specified on inner and outer walls, taking account of the temperature dependence of the thermophysical properties of fluid. The perimeter average Nusselt numbers and dimensionless average axial velocities in the ducts of various cross sections are well correlated by usingξ as a characteristics length ; and the Nusselt numbers of the annular ducts show very good accordance with those of parallel plates rather than those of a circular duct. A general representation of the average Nusselt numbers is presented for wide ranges of the radius ratio and Prandtl number. When the fluid physical properties are evaluated at the average temperature of the fully developed flow, the variable property solutions correlate with the constant property solutions.

View full abstract

Heat flux from the human skin to the environment was measured with heat flux transducers over the entire skin surface of a male subject. The total heat flux for bare skin was 113 W and when the subject wore a loose sports shirt and trousers was 84.9 W, in a room at 20°C. Heat flux distribution was found to be approximately symmetrical from the centerline of a torso without clothes. An approximate correspondence with skin surface temperature was obtained. For comparison, a constant temperature bath made of stainless steel plates was constructed in the shape of a human body. The total heat flux from this human model, measured by the previous transducers, agreed fairly well with the total electric energy supplied to the constant temperature bath.

View full abstract

By adopting an appropriate near-wall function in a K-ε model, turbulent convective heat transfer in the 90^o-bend tube with a straight tail section has been analyzed numerically for a low-Reynolds-number turbulent-flow regime. The results are presented in the form of velocity profile, temperature distribution, turbulent intensity distribution, friction factor and heat transfer for the condition of a heat flux constant. The proposed model of the near-wall function describes well the low-Reynolds-number turbulence field including a reverse transition mechanism.

View full abstract

Experimental measurements are compared with theoretical predictions for heat and mass transfer from hot and highly humid air (130°C, 0-15 wt% of water vapor content) to horizontal cooling tubes (30°C). The Reynolds number range is 2 800-9 000. Heat- and mass-transfer coefficients are measured under conditions with and without condensation present by using a newly developed condensate collector. It is found that the convective heat transfer and mass transfer are enhanced by 68% and 28% respectively by increasing the water vapor content in the air flow from 0 to 15 wt%. Mass transfer data fits well with Rose's equation and heat transfer data are shown to give 20-30% higher values than the prediction. This means that the analogy between heat and mass transfer does not hold under condensation conditions. This phenomenon is explained qualitatively by considering that the water vapor is consumed faster than the thermal energy in the separated flow region on the rear side of the tubes, due to the value of Pr/Sc being greater than unity.

View full abstract

Soot, total hydrocarbon, oxygen and the temperature of turbulent diffusion flames together with partially premixed turbulent diffusion flames of C₃H₈ and C₂H₂, were measured under the normal temperature and pressure to study the soot formation process in a flame. To study the effect of fuel properties on the soot formation process, the soot concentration in CH₄, C₂H₂, C₂H₄, and C₂H₆ flames was also measured. The soot concentration is a flame increased along the flame axis as long as the oxygen concentration was low and the total hydrocarbon was not consumed. If the total hydrocarbon was consumed or oxygen increased, the soot concentration decreased. The maximum soot concentration in a flame was increased by increasing the flame temperature in the soot formation zone in the flame. However, if the temperature was increased beyond 1 400 K, the maximum soot concentration decreased. The maximum soot concentration was also increased by increasing the C/H ratio of the fuel.

View full abstract

The methanol oxidation reaction on the catalysts of Pt, Pd, Rh, Ir, and Cu has been investigated to select suitable catalysts for the purification of methanol engine exhaust gases. For all of these catalysts, the presence of NO showed an inhibitory effect on the methanol oxidation reaction and caused the formation of formaldehyde. By comparison with other exhaust components (CO, H₂O, HC), it was found that NO had a stronger effect than those components. In the presence of NO, the relative activity sequence was also obtained as Pt≒Pd>Rh≒Ir>Cu. The temperature conditions for formaldehyde formation were revealed for each catalyst in the presence of NO, and it was also found that formaldehyde formation was suppressed the most by using the Pd catalyst. Consequently, Pd is a catalyst which is superior for reducing unburned methanol.

View full abstract

The final goal of this study is to develop a multipurpose measurement system, which can simultaneously obtain thermal conductivity, total hemispherical emittance, thermal expansion, thermal diffusivity and specific heat by a direct electrical heating method. In this report, the principle of the newly-developed method for measuring thermal diffusivity is described. Rod samples of silver (purity 99.99%) and austenitic stainless steel (JIS SUS 304) of 200 mm in length and 1.494 mm in radius are used in order to verify the applicability of this method. Theoretical evaluations of errors in this measurement are also conducted. From the experimental results and their evaluation, it becomes clear that this method is satisfactory for precisely measuring the thermal diffusivity of a rod sample.

View full abstract