In this paper, a novel and effective numerical solution procedure for solving convection-dominated flows is proposed in the context of the finite element method (FEM). The procedure, which we call CIP-FEM, is a combination of the cubic interpolated propagation (CIP) method in the advection phase with FEM in the diffusion phase. Applicability of the method is shown in the comparison of other existing solutions for the rotating cone problem known as a bench mark test in a two-dimensional convective diffusion equation.

View full abstract

In the present paper, unsteady natural convection flows of air in a square cavity with differentially heated walls are investigated numerically using the hybrid-GSMAC (generalized and simplified marker and cell) finite-element method. A direct-numerical simulation of two-dimensional turbulence is shown in the present paper without introducing any random forces. The present result at Rα=1.0×10⁸ is compared with the other numerical results. The present result of calculation is in good agreement with that of the others This study allows us to say that the hybrid-GSMAC method suggested by Tanahashi is clarified. In addition, selected frames from an animation generated from the computational results at Rα=1.0×10⁹ show very clearly the formation of large-scale eddies via the following sequence : initial instability, proceeding through transition, and eventually the statistical steady state. Using this method, it is possible to investigate natural convection and heat transfer phenomena at high Rayleigh numbers up to 10¹0.

View full abstract

A new type of sheath flow chamber with low pressure loss is developed using micro machining. The chamber is formed by laminating three l00-μm-thick, photo etched, metal plates. The device uses sheath-flow geometry ; particles in suspension, which enter the chamber via an axial specimen nozzle, are enveloped by a buffering sheath-flow and transported to a capillary tube with cross section of 300μm×300μm. A finite-element viscous-flow analysis is used to examine the flow behavior in the quasi-two-dimensional passage configuration to which produces fluid-dynamic focusing flow. Experimental measurement shows a smooth constricted sheath-flow and pressure loss was reduced to one-tenth in comparison with the conventionally used glass flow chamber.

View full abstract

The κ-εmodel has become the most popular of the turbulence models and is often used in many calculations of flow of practical interest. Various low Reynolds number κ-ε models have been proposed ; however, they are not able to correctly represent the budget of the turbulent energy equation near the wall. This paper presents an improved two-equation model with a correct asymptotic near-wall behavior. A new equation of ε^^[=ε-2ν(√∂<κ>/∂y)²] is proposed instead of the conventional one for ε itself. Moreover, the pressure diffusion π^*_κ is introduced in the κ equation and a new term-ε^^ε^^/κ is employed in the ε^^ equation. The improved model is tested against the channel flow, the pipe flow and the channel flow with injection and suction on wall surfaces. The results are compared with the ones obtained by direct numerical simulation (DNS). It is found that the present model reproduces well the turbulent quantities obtained by the DNS.

View full abstract

Synchronized vortex shedding from a rotating regular polygonal prism in a uniform flow is investigated numerically and experimentally The Karman vortex in a wake behind the prism rotating in a wind tunnel is investigated by using a hot-wire anemometer. The experimental results show that the Karman vortex shedding is synchronized with the rotation of the prism for a wide range of spill parameters. In numerical analysis, a body-fitted grid system that coincides with a moving boundary of the rotating prism is installed and the 2-dimensional Navier-Stokes equation is solved using the finite difference method. The numerical results show good agreement with the experimental ones. It is found from the experimental and numerical results that the general relation-ship between spin parameter S and Strouhal number St for an N-sided rotating polygonal prism is St=N/(κπ)·S(κ=1, 2, 3···).

View full abstract

The application of large eddy simulation (LES) to practical engineering problems has been attempted by many researchers. In recent years, flow fields that have not only simple geometry but also complicated flow structures have been simulated by LES. In this paper, turbulent flow around a stationary and an oscillating rectangular cylinder at high Reynolds numbers of 2.2×10⁴ and 7.14×10⁴ has been analyzed by LES. The main objectives of this work are to compare computed results with experimental values and to prove the applicability of LES to flow induced vibration problems. Drag coefficients, base pressure coefficients and Strouhal numbers were in fairly good agreement with experimental values, while the classical Smagorinsky model was used as a subgrid scale (SGS) model. In cases of flow around an oscillating rectangular cylinder, we successfully simulated the lock-in phenomenon whereby the vortex-shedding frequency equals the oscillating frequency. The width of the lock-in region and phase angles between cylinder displacement and lift coefficient at various oscillating frequencies were in good agreement with experimental results. Both lock-in and nonlock-in states of flow were observed in the vicinity of the boundary frequency of the lock-in region.

View full abstract

Fine-scale structure and local isotropy were experimentally investigated for turbulence fields of turbulence Reynolds numbers of R_λ=80∼393. Waveform analysis was performed for velocity fluctuations independently in the low wavenumber range, the inertial subrange and the viscous dissipation range and also for the time derivative of the velocity fluctuations. For R_λ>200, the turbulence fluctuations were random in the low wavenumber range but intermittent in the other two wavenumber ranges. Probability density analysis showed that the flatness factor was about 3 in the low wavenumber range and the intermittency was stronger in the viscous dissipation range that in the inertial subrange. The flatness factor of the time derivative of the velocity fluctuations monotonically increased with R_λ and reached about 7.08 at R_λ=393. We introduced the anisotropic tensor spectrum. The scale of the maximum isotropic eddy, l_G, determined the upper limit of the inertial subrange where the turbulence eddies smaller than l_G attained local isotropy and universal equilibrium in the energy cascade process.

View full abstract

This paper presents a numerical analysis to investigate three-dimensional turbulent structure and fluid flow behavior in a rectangular channel flow with streamwise-periodic ribs mounted on facing walls. In the calculation, an algebraic Reynolds stress model together with boundary-fitted coordinate system is applied to a periodically ribbed channel in order to solve the accelerated, separated and recirculating flows. The calculated results are compared with the experimental results available to examine the validity of the present method. As a result of this calculation, it was found that the present method predicted well the distributions of mean flow velocity and reattachment length. Moreover, the calculated results of turbulent energy and shear stress are in good agreement with the experimental data. Although it seems that the convection term of the Reynolds stress equation has an effect on the distributions of Reynolds stresses in this periodically ribbed channel flow, an algebraic Reynolds stress model, in which convection and diffusion terms are transformed from the differential form to algebraic form, predicts the experimental results of Reynolds stresses with little discrepancy. As regards the secondary flow, calculated flow patterns differ with the experimental results. In calculation, large variations of secondary flow direction are predicted along the streamwise direction whereas the experimental results show an unchanged flow pattern in both regions with and without recirculating flow.

View full abstract

To date the Rankine vortex or the viscous vortex has been used for the vortex method. Since these vortices do not have actual turbulent characteristics, a new vortex model is required for turbulent flow analysis. In this study, we proposed a turbulent vortex model including logarithmic circulaticm region and applied it to predict the growth of a large-scale turbulent vortex. The present turbulent vortex model is proven to predict the turbulent vortex structure, while neither the Lamb vortex model nor the Squire vortex model are necessarily appropriate.

View full abstract

This paper discribes a method of predicting hemolysis induced by turbulent shear stress (Reynolds stress) in simplified orifice pipe flow. In a centrifugal blood pump, serious hemolysis occurs at the impeller or the casing edge. The turbulent shear stress computed using a low-Reynolds numberκ-ε model near the edge proves to be several hundred times the laminar shear stress (molecular shear stress). The peak turbulent shear stress is about 2500Pa, which is much larger than that obtained from a conventional hemolysis test using viscometer. Then the threshould of turbulent shear stress for hemolysis is determined by comparing with that obtained in hemolysis experiments of pipe-orifice blood flow.

View full abstract

The present paper deals with experimental investigation using micron-sized distributed surface geometry on a flat plate in order to observe effective drag reduction. Namely, the objective is to verify Tani's hypothesis that distributed rough surface can even reduce turbulent drag in a certain Reynolds number range. So, we directly measured the net drag of a flat plate wrapped by a sheet with micro-sized distributed rough surface geometry. Obtained data showed that a limited amount of drag reduction is observed in the case of certain kinds of surface geometries. This result requires a new explanation of how the turbulent energy is suppressed in the turbulent boundary layer, since existing turbulent drag reduction mechanisms for a riblet surface is always explained together with streamwise groove structure. Therefore, by analyzing the obtained results, a more appropriate energy production mechanism in the turbulent boundary layer might be found out. The present experiment is the first step in a series of investigations concerning the above title.

View full abstract

Properties of light emission from a cavitation generated by a circular orifice shear flow have been investigated experimentally. In order to enhance the weak emission from clear water cavitation, xenon gas and luminol were dissolved in the water. The emission intensity was directly measured by an optical system with a photon counter and a spectrometer. It was seen that the emission could be enhanced by more than 100 times, and that the measured spatial profile of emission and its spectrum for xenon-water cavitation were very similar to those for clear water. The spectrum for luminol-water cavitation was a simple superposition of the spectrum from clear water and that due to luminol. The color temperature of emission from xenon-water cavitation was as high as roughly 20 000K.

View full abstract

An Experimental investigation was carried out on air entrainment into water flowing downward in a vertical pipe. Local flow rates of air and water in a fluid layer of annular water flow, formed on the pipe wall, were measured precisely by using a small tube probe. The distributions of local flow rates in the radial direction of the pipe indicate that the fluid layer consists of a water layer adjacent to the pipe wall and a water-air layer located inside of the water layer. The water-air layer is formed as a result of air entrainment. The amount of entrained air increases with the increase of the axial distance of the pipe and of the total flow rate of water.

View full abstract

Gas ingestion phenomena occur at a gas-liquid (air-sealant) interface in a viscoseal, which is a kind of noncontact seal, and the ingested gas adversely affects the performance of the viscoseal. A new vacuum pump taking advantage of the gas ingestion phenomena has been proposed by the authors. In this paper, the characteristics of the gas ingestion phenomena are described and the performance of the proposed vacuum pump is discussed with reference to the experimental results. Experiments were carried out using an apparatus with a threaded shaft (480-4300rpm) with 40mm diameter, 145mm length and 0.2mm clearance between shaft and housing. Water and silicone oilwere used as sealant.

View full abstract

A study on the effects of image gray scale and pixel number on image quantification was performed analytically and experimentally in order to establish a method of measuring multiphase flows using neutron radiography. It was found from a simplified model that the resolution of an object thickness to be measured had logarithmic characteristics for the object thickness, and there is a possibility that the measured void fraction in high- and low-void-fraction regions might include large error. It was shown from the numerical analysis that the measurement error would be within 5% except for the annular flow region with the void fraction greater than 90%, when void fraction of an air-water two-phase flow in a channel with the local gap smaller than 9mm was measured under the condition of the image gray scale=100. It was revealed from a simplified model that the error due to the number of pixels would be within 5% when the number of pixels projecting the object is ten times greater than that projecting the object boundary.

View full abstract

When a coiled pipe rotates about the coil axis, the effect of rotation interacts with centrifugal and viscous effects to complicate the flow characteristics beyond those usually seen in stationary curved pipes. Following previous papers on the flow and heat transfer characteristics for flows with large Rossby number, the effects of small Rossby number on the characteristics are studied for fully developed laminar flows through computational studies.

View full abstract

Fully developed laminar flow in a square duct rotating about a perpendicular axis is investigated by similarity arguments and computational studies. Flow and heat transfer characteristics are expressed by the dimensionless number K_LR which was introduced by the author, and the Rossby number R_O. The flow patterns are shown for large and small values of K_LR and R_O. It is shown that, above a critical K_LR, a pair of vortices due to hydrodynamic instability appears near the center of the pressure-side wall in addition to the secondary flow vortices. The computational results for friction factor and heat transfer are given and compared with those of flow in a circular pipe. The properties of secondary flow and flow instability are discussed in some detail.

View full abstract

A small rod was set upstream of a circular cylinder for enhancement of heat transfer and reduction of drag of the circular cylinder. The diameter of the cylinder was 40mm, and that of the rod d was varied from 1 to 12 mm. The distance between the axes of the cylinder and rod L was varied from 50 to 120mm, and Reynolds number ranged from 1.5×10⁴ to 6.2×10⁴. The local heat transfer distribution has a maximum at the angle of ±30∼35°from the front stagnation point, regardless of the presence or absence of vortex shedding from the rod. Therefore, the heat transfer at the front face increases markedly. Beyond R_e≥3.1×10⁴, a second maximum of heat transfer appears at φ=±11O°upon transition to turbulent flow. In the separated region at the rear face, the heat transfer decreases. The overall heat transfer increases with an increase in d/D and a decrease in L/D. This method is useful for higher Reynolds numbers. The optimum conditions are d/D=0.25 and L/D=1.25, that is, the case in which quasi-stationary vortices are formed between the rod and the cylinder. In this case, the overall heat transfer increases by 40%, and the total drag coefficient included the drag of the rod is reduced by 43% compared to that of without the rod.

View full abstract

A new cross-correlation method, which is called a wavelet cross-correlation analysis and expresses the statistical cross-correlation of any signal in terms of period and time delay, is proposed and its main properties are presented. By analyzing two typical signals, the cross-correlations of components with different periods can be detected, and it is shown that wavelet cross-correlation can overcome the limitations of classical cross-correlation. As a practical application to the field of fluid mechanics, wavelet cross-correlation is employed to analyze cross-correlation relationships between χ-components of the fluctuation velocities at two points on opposite sides of the centerline and along the centerline of a plane turbulent jet in terms of period and time delay. From the distributions of the wavelet cross-correlation coefficient, similar structures of motions with various scales are shown, and the successive branching of vortex structures is observed. The period of vortex and apparent flapping motions can be easily determined in terms of period and time delay.

View full abstract

An axisymmetric gas-jet impinging onto a solid surface has been studied experimentally. Mean and fluctuating velocities are measured by the technique of rotating a probe with an inclined hot wire. Wall shear stresses are measured using a glue-on probe. Figures show distributions of the mean velocity, the turbulence energy, the Reynolds stress and the static pressure, and mean and turbulence characteristics of the jet are clarified. By comparing with the jet impinging onto a solid flat surface, the influence of curvature of wall is examined.

View full abstract

It is of interest to clarify dynamic properties of water jet pressure acting on the surface of an object to be cut in order to clarify the cutting mechanism. This pressure includes a wide range of frequency components from the pulsation of a plunger pump to the impact pressure caused by the impingement of droplets. In such case, the transducer characteristics affect the detected wave considerably, and the source wave of the pressure differs from the detected wave. In this paper a method of determining the source wave of stagnation pressure is presented. In addition, the relation-ship between the characteristics of the stagnation pressure of a water jet and the flow structure of a water jet is examined by flash photography using pulsed laser light. As a result, it is clarified that the dynamic property of stagnation pressure is closely related to the wavy motion of a water jet.

View full abstract

An analytical study has been conducted on steam injector systems (SISs) for next-generation nuclear reactors. The steam injector is a simple, compact passive steam jet pump. It is easy to start up without electric power. A characteristic analysis model for a steam injector has been developed based on experimental data. To evaluate steam injector characteristics, parametric analyses were carried out. The analytical results indicated good performance of the SIS, that it functions over a very wide steam pressure range, and that it is suitable for next-generation nuclear reactors.

View full abstract

The purpose of this experiment is to reduce edge tone generated by flow of a jet-wedge system. An attempt to reduce the edge tone has been carried out by putting sidewalls parallel to the jet axis. The edge tone has been reduced by the use of sidewalls even though these are much shorter than the edge distance. The mechanism of the edge tone reduction has been studied using sidewalls which have slits. It has been found that the edge tone is reduced when the oscillation of the jet is suppressed by decreasing the entrainment of ambient fluid near the nozzle exit.

View full abstract

Compression waves generated by a high-speed train entering a tunnel were measured in three Shinkansen tunnels and the results were compared with the numerical values calculated using the TVD scheme. The strength of a compression wave is exponentially attenuated with distance as it propagates along the tunnel in both slab and ballast track tunnels, and the attenuation in the ballast track tunnel is considerably larger than that in the slab track tunnel. In the slab track tunnel, the compression wave is steepened as it propagates, while it spreads in the ballast track tunnel. The criterion for judgment whether the compression wave is steepened or spreads is expressed by the value of acoustic Reynolds number of about l0∼15.

View full abstract

A screw propeller is the most typical mechanism for ship propulsion. However, it is well known that some species of aquatic animals such as tuna and dolphin are able to swim at a high speed by powerful fanning of their caudal fins. The swimming motion of these animals is considered to be a combination of the heaving and pitching motion of the tail fin (the oscillating wing). The authors have developed a two-hinge oscillating fin propulsion mechanism which basically follows the swimming motion of aquatic animals'. The propulsion system is driven by a 13 PS motorcycle engine mounted on an experimental ship. The ship is 3.6m in length, 1m in width, and about 160kgf in weight. The wing is flat plate of 3.2mm thickness and 0.27m² area, and its aspect ration is 6.3. The wing motion is determined by the balance of forces between a flowing fluid and the torsion of a coiled spring attached to the 2nd hinge at each engine frequency. The maximum ship speed of l.58m/s is obtained in the case of 3.53N·m/deg spring strength and 2.8Hz frequency.

View full abstract

The leak flow through radial and axial clearances between fixed and orbiting scrolls is considered the dominant factor which determines the pumping performance of a dry-scroll vacuum pump. Since the leak flow sometimes ranges from viscous to slip to molecular flow along a clearance, an equation which is able to describe the flow through the three flow regimes is developed by the weighted linear combination of two equations for molecular flow and slip flow. The pressure distribution in gas pockets is considered another important factor for the pumping performance, which is calculated in the same manner as the leak flow. The predicted ultimate pressures show satisfactory agreement with the ultimate pressures of a practical machine.

View full abstract

The purpose of this study is to investigate the pressure effects on the entrainment flow rates in vertical gas-liquid annular two-phase flow. The cross-sectional entrainment flow rates were measured using a isokinetic probe method. It was found that the behavior of cross-sectional entrainment flow rate profiles are divided into low-and high-pressure regions. Also, the entrainment flow rates amount to 90% of the total liquid flow rate under high-pressure conditions, In this study, system pressure in the closed-loop system was changed substantially from 0.3 to 20MPa at a constant fluid temperature in vertical upward flow.

View full abstract

This report provides an assessment of five correlations for estimating the entrainment flow rate. These correlations are compared using air-water experimental entrainment probe data obtained under a wide range of system pressure from 0.30 to 20MPa. There are no correlations that are able to predict the pressure effect under a wide range of pressure for the entrainment flow rate. Two new correlations for the pressure ranges 0.12∼5.0MPa and 7.0∼20.0MPa for predicting the entrainment flow rate are presented. Also, the correlations for minimum superficial gas velocity that be able to divide between liquid lumps and entrainment flow rates are presented, and these conditions are coincide with the mean film thickness 1.Omm.

View full abstract

The authors investigate the behavior of a bubble which bridges a gap in a cross section of a horizontal narrow divergent passage under the Earth's gravity condition (1G). In a narrow passage, inertia forces are known to be small compared with viscous forces. Also, gravity force is not dominant for bubble behavior in a horizontal narrow passage. In this sense, the bubble behavior in the passage is similar to that under a microgravity condition. It is important to understand the bubble behavior in relation to separating gas from a gas-liquid two-phase flow and controlling a gas-liquid interface under a microgravity condition. A one-dimensional momentum equation for the bubble behavior was derived. The equation was converted into an ordinary differential equation with respect to the upstream interface and was solved. Analytical results were compared with experimental data. As a result, effects of gap size, bubble projected area, and divergent angle on the bubble behavior were explained qualitatively.

View full abstract

Single-drop experiments are carried out with the object of studying the propagation process of spontaneous vapor explosion and the behavior of pressure waves induced by the explosion itself. Both a molten tin drop/water system and the molten LiNO₃ drop / ethanol system are examined. Propagation of explosion between two tin drops is confirmed in tin / water experiments, though the apparent propagation velocity is only about 5m/s and the explosion behavior differs largely between two drops. In the LiNO₃ / ethanol system, the traveling behavior of the pressure wave in a one-dimensional channel is analyzed and the velocity through the interaction zone is measured. Full time coherence is confirmed in a 25-drop experiment and it is revealed that the velocity depends largely on the development stage of the vapor explosion.

View full abstract

Critical heat flux (CHF) of a uniformly heated tube usually occurs at the tube exit but is sometimes detected at an upstream location under high mass flux at high system pressure. To clarify the characteristics of upstream CHF, systematic experiments were conducted using R115 as the test fluid over wide experimental ranges of system pressure, mass flux, tube length and liquid inlet subcooling. As the inner surface roughness of the test tube was found to have a strong influence on the CHF, three test tubes with different inner surface finishes were used. From the experiments, the followings results are obtained : (1) upstream CHF is an autonomous phenomenon ; (2) up-stream CHF is likely to take place in a rough tube ; (3) upstream CHF is more likely to take place at low inlet subcooling and high pressure under high mass flux ; (4) upstream CHF is greater than usual downstream CHF for the same inlet subcooling ; (5) downstream CHF can be correlated uniquely as a function of local quality whereas upstream CHF data deviate from the unique relation.

View full abstract

Heat transfer coefficients in nucleate pool boiling of five binary mixtures at atmospheric pressure were measured on an upward-facing circular plate for a wide range of heat fluxes from about 15percent of the critical heat flux (CHF) to close to CHF. As has been observed in many previous measurements, heat transfer coefficients were reduced in comparison with the interpolated values between their constituent pure components. This reduction was dependent on the mixture composition and became more pronounced with increasing heat flux. To provide the simplest and most reliable prediction of the reduction of heat transfer coefficients regarding the mixture composition and heat flux, the Thome correlation was modified so as to include the effect of heat flux in a dimensionless form. The developed correlation succeeds in correlating the present experimental data within ±20 percent accuracy.

View full abstract

Both heat and mass transfer in the gas phase and heat transfer in the liquid phase are examined experimentally for film condensation of organic binary mixtures such as ethanol-water and methanol-water. Experimental results on average heat flux, gas-liquid interface temperature and liquid-phase Nusselt number are compared with the analytical solutions based on a stagnant film theory and heat transfer relationships for film condensation from a pure vapor. Experimental results on heat transfer agree well with the analytical solutions, except that the liquid-phase Nusselt number in experimental results in the condition of low mass fraction of water is considerably higher than that in analytical solutions. This high value of the liquid-phase Nusselt number is considered to be caused by dropwise condensation in liquid phase. However, its effect on the tube bundle is not so remarkable compared with that in gravity-controlled condensation on a vertical surface, This is considered to be caused by the condensate inundation effect.

View full abstract

The characteristics of heat exchangers with offset-type plate fins for space stations are studied for Reynolds number less than 300 based on the hydraulic diameter. Three-dimensional analysis is carried out to study the effects of the following parameters on the heat transfer and the flow characteristics : (a) the thermal boundary layer developing on the bottom plate and on the fins on the plate, (b) the aspect ratio (height / pitch) of the cross section of the flow passage, the fin thickness, the fin length in the direction of the flow, the thermal conductivity of the fluid and the fins, and the Prandtl number of the fluid. The results obtained are as follows. (1) The heat-transfer coefficient on the fin surface is characterized by the thermal-conductivity ratio of fluid to fin material. When the thermal conductivity of the fin material approaches that of the fluid, the heat-transfer coefficient on the fin surface becomes low. (2) The optimum condition of the aspect ratio depends on the value of the thermal-conductivity ratio between the fluid and the fins. (3) When the aspect ratio becomes large or small, the friction factor of offset fins approaches that of fully developed duct flow with the same aspect ratio as the Reynolds number decreases.

View full abstract

Effects of pulsation on flow and heat transfer characteristics are experimentally examined in the pulsating pipe flows having sinusoidal velocity fluctuation around a nonzero mean. By systematically varying three pulsation parameters (the amplitude, frequency, and mean velocity), time-averaged and fluctuating temperature profiles are measured under the heating condition of constant wall temperature using saturated vapor. The mean Nusselt number, Nu_p, is calculated, and compared with that in ordinary turbulent pipe flows without pulsation. The results show that Nu_p decreases initially as the pulsation amplitude increases, then recovers gradually, and finally becomes much greater than the original value. In the pulsating pipe flows with a non-zero mean velocity, therefore, pulsation cannot always promote heat transfer, but sometime suppresses it, depending mainly on the pulsation amplitude and mean velocity. It is also found that these heat transfer characteristics of a pulsating pipe flow are controlled by the transition of flow patterns with pulsation amplitude from a fully turbulent flow to a conditionally turbulent flow via a transitional flow.

View full abstract

Heating a cylinder should affect the flow pattern around it and the heat transfer from its surface. Numerical computation of the flow and related heat transfer has been carried out for a circular cylinder mounted in a channel flow for a free-forced combined convection regime. In the case that buoyancy assists the flow, the results showed that the transition from unsteady flow to steady flow occurred under different conditions depending on the Ri number, Re number and blockage ratio. With a constant Re number, the critical Ri number at which the transition occurs decreases with increasing blockage ratio, and with a constant Ri number, the transition Re number increases with the blockage ratio. In the case of negative buoyancy, on the other hand, the increase of the Ri number facilitates the growth of the length scale of the Karman vortices. The buoyancy effect on flow instability is less conspicuous compared with that in the above two cases in the case of horizontal flow, but local heat transfer characteristics are still changed by the buoyancy effects.

View full abstract

To control the flow around a flat plate, a small rod was set upstream of the plate. The chord length of the plate, D, was 50mm and the diameter of the rod, d, and the distance between the axes of the rod and plate, L, were varied. The Reynolds number ranged from l.3×10⁴ to 7.7×10⁴. For the case without vortex shedding from the rod, the shear layer from the rod reattaches on the front face of the plate. Consequently, quasi-stationary vortices are formed between the rod and the plate. In this case, at d/D=0.4 and L/D=1.4∼2.0, the maximum reduction of total drag coefficient is 20∼30%, the average heat transfer on each face increases and the overall heat transfer increases by about 40∼50% compared with values obtained without the rod in place.

View full abstract

Experiments were conducted to measure the local heat transfer on an endwall with a pin fin array. Heat transfer behavior was examined for the cases of a single pin, a single row, in-line and staggered arrays having six streamwise rows. Thermosensitive liquid crystal film was used to measure the local heat transfer coefficient on the endwall. Neural network was applied for the color-to-temperature transformation of the thermosensitive liquid crystal. Local heat transfer on the endwall having a single row of pin fin was affected by flow acceleration between the pin fins rather than the horseshoe vortex around the pin fin. Therefore the average Nusselt number exhibited a good correlation with the Reynolds number Re_max which was based on the average velocity of the minimum flow area, regardless of the pin fin spacing. For in-line and staggered arrays, the average Nusselt numbers associated with the Reynolds number Re_max decreased with the reduction of the pin fin spacing.

View full abstract

Heat transfer and turbulence characteristics in the stagnation of an excited plane impinging jet have been investigated experimentally. Jet excitation was utilized to lock the phase of vortex motion and temperature variation of both fluid and wall. The velocity and temperature field were measured by digital particle image velocimetry (DPIV) and laser-induced fluorescence (LIF). In the mean flow field, the excitation makes closer the location of the impinging plate which has a maximum heat transfer coefficient. The local Nusselt number with excitation along the stagnation line is larger than that without excitation for an impinging plate set at a distance of 5-6 times the nozzle width, where the intensity of the fluctuating velocity normal to the wall and turbulent heat flux component normal to the wall are also increased. The turbulent heat flux and Reynold's stress that are mainly generated by counter-rotating vortex pairs contribute to reduce velocity and thermal boundary layer thickness. When the spanwise vortices arrive in the stagnation region, vorticity of the counter-rotating vortex pairs is intensified, and this enhances the intensity of the turbulent heat flux and Reynold's stress which causes high heat transfer instantaneously.

View full abstract

The objective of this study is to evaluate simultaneously the time dependence of the thermal diffusivity of carbon-carbon composites (C/C composites) and their porosity during heat treatment using the electromagnetic ultrasonic technique. The present report describes two kinds of experiments conducted to confirm the principle for simultaneous measurement of both the ultrasonic velocity (used to evaluate the porosity) and the thermal diffusivity at room temperature. For each material, the sample used in both experiments was identical. The ultrasonic velocity of type 304 stainless steel and its thermal diffusivity were 5.85km/s and 3.8mm²/s with precisions of ±1.6% and ±8%, respectively. The ultrasonic velocity of a two-dimensional woven C/C composite and its thermal diffusivity were 2.86km/s and 4.8mm²/s with precisions of±5.0% and ±8%, respectively. The results appear to indicate that the electromagnetic ultrasonic technique can measure the ultrasonic velocity and the thermal diffusivity simultaneously and that it is also applicable to C/C composites for the measurements.

View full abstract

We propose that an increasing value in entropy inequality is not a local value of entropy but an average value in a mesodomain. From this standpoint, new balance equations are obtained by applying the concept of generalized continuum mechanics to mass transfer. In the present paper, it is clarified that the influence of microscopic gradient of mass concentration should be included in Fick's first law and Fourier's law. Even if thermal effects on diffusion are neglected, the diffusion equations obtained here are simultaneous differential equations with two undetermined values, which coincide with the conventional diffusion equation in a special case. The conventional diffusion equation describing infinite velocity of propagation associated with diffusion entails a contradiction. The velocity of propagation defined here is shown in the results of numerical analysis for an extreme initial state. Consequently, it is indicated that the present theory gives an acceptable solution to the above problem of the conventional theory.

View full abstract

In this report, a simple model to predict heat transfer distribution in the high temperature region of spray cooling is developed by focusing on the effect of sensible heat of droplets. In the model, it is assumed that the droplets rebound repeatedly on the surface and heat transfer upon droplet impact is proportional to the sensible heat which heats up the droplets to the saturation temperature. The calculated results are compared with existing experimental results and it is shown that the present model can predict the heat transfer distribution in the high temperature region of spray cooling.

View full abstract

We present experimental results on heat transfer distribution in the high temperature region of spray cooling interacting with subcooled liquid film flow. The results show that the flow field can be divided into the interacting and film flow regions by heat transfer distribution. In the interacting region, the heat transfer coefficient can be correlated to the liquid-film-flow heat transfer by using a heat-transfer enhancement coefficient defined as the ratio of the droplet flow rate to liquid film velocity. In the wall region, it can be predicted from the equation obtained from the previous study, which is very similar to that of turbulent heat transfer of single-phase flow.

View full abstract

Generally speaking, it is very difficult to clarify the growth process of large-scale vortices in a transient gas jet. The vortices have a great effect on the mechanism of mixture formation between the jet itself and the surroundings. The objective of the study presented here is the clarification of this mechanism in both a free jet and a jet impinging on a flat wall by means of the flow visualization of 2D images obtain by a thin sheet of laser light in experiments and numerical analysis by the discrete vortex method. The mechanism of the vortex growth and the coherent structure of vortices entraining the surroundings are clarified in the case of a free jet. In the wall impinging jet, the structure inside the jet is divided into three regions, namely, free jet region, impingement region and wall-main jet region. The results of the numerical analysis agree qualitatively with the experimental data.

View full abstract

Using the bypass controlling method, the outlet temperature from an energy storage tank is obtained as required. In this paper, a numerical method simulating the response of the storage tank is proposed. The models show good agreement with experimental data. Then the performance is estimated by the amount of energy necessary to obtain the required temperature. The efficiency is shown as a function of the total flow rate and the required temperature. The effects of the capsule array and capsule size on the performance are discussed. Finally the effects of the heat transfer promotion of the inside and/or outside of the capsules on the total performance are discussed.

View full abstract

One means of improving the efficiency of waste-to-energy (WTE) systems is the use of combined cycle WTE (CC-WTE) systems that have a gas turbine and a steam turbine. On the other hand, steam injection to a gas turbine combustor has been used as a means of raising the power output and of improving the thermal efficiency. It is considered possible to realize a high-efficiency WTE system by injecting steam produced by the waste heat boiler of a waste incinerating plant as in a CC-WTE system. We describe the performance analysis of steam-injected gas turbine (SIGT) WTE system in terms of the repowering efficiency defined in this paper. As a result, it is concluded that the repowering efficiencies of SIGT-WTE systems are superior to those of CC-WTE systems in which utilize high-pressure-ratio of gas turbines.

View full abstract

The capillary pump loop (CPL), which employs the latent heat of vaporization and condensation and does not need a driving source, is considered as a promising candidate for space thermal control systems because of its high reliability and lightness. In this study, the performance and characteristics tests using a prototype CPL model which has two flat-type evaporators and a reservoir as an operational control device was made. To withstand the wide range of heat loads in space applications, two tests, one conducted under an even heat load and the other under an uneven heat load, were performed. In these experiments, our interest was focused on the region where we can get stable operations, heat transfer mechanisms at the evaporator section and controllability of the reservoir. In order to investigate in greater detail the flow behavior of working fluid inside the evaporator section, visual observation was also carried out using a transparency loop. As a result, an important fundamental data base required for the establishment of the theoretical prediction of CPL performance was obtained.

View full abstract

The capillary pump loop (CPL), which employs the latent heat of vaporization and condensation and does not need a driving source, is considered as a promising candidate for space thermal control system because of its high reliability and lightness. Based on the experimental data base using a prototype capillary pump loop which has two flat-type evaporators and a reservoir, a new theoretical model for predicting CPL performance was developed in this study. In this theoretical model, the receding process of the vapor-liquid interface observed inside the evaporator was taken into consideration in order to make a more precise prediction. Because a successful agreement was found between the theoretical prediction and both the fundamental perfomance data base and the additional experimental data obtained using various combinations of wick size and working fluid, the validity of this theoretical method was confirmed.

View full abstract

A linear analysis model for a free piston Vuilleumier heat pump (FPVM) was constructed to derive closed form solutions for the operating frequency, phase angle and displacer strokes at self-excited operation and forced oscillation. Using the linear analysis, conditions for self-excited operation and the effects of connecting the hot displacer and the cold displacer by a spring on the dynamic motion of the displacers and the FPVM performance were investigated. Energy flow in the FPVM was also obtained. A simple 2nd order isothermal simulation combined with the linear analysis was developed which enabled culculation of the FPVM performance characteristics. It was found that the simulation results agreed well qualitatively with results of experiments at self-excited operation. The simulation was used to calculate the dependences of the dynamic motion of the displacers and the FPVM performance on the spring constants, displacer masses and damping coefficients.

View full abstract

Apremixed-combustion-type boiler with a multiport-type burner has many advantages in terms of NO_x exhaust, thermal efficiency, furnace design, and controllability of energy supply. Further-more, both the noise of the combustion and that of mechanical assembly are expected to be small in comparison with the noise of a diffusion combustion burner. However, the designed boiler has higher capability of exciting the combustion-driven oscillation noise caused by interaction among the components, and it is difficult to determine a systematic solution to this problem in advance. As a study on the mechanism of combustion oscillation and the noise reduction method, based on the dynamic analysis of acoustic behavior of flow heated by a flame, the basic characteristics of oscillation in an actual boiler and a method of modeling it are reported. The excitation condition and the influence of dynamic behavior of the flame on the noise excitation are clarified.

View full abstract