It was found in the preceding study that the boundary layer separation over a circular cylinder seems to have three-dimensional separation characteristics and its three-dimensionality appears to be related to the three-dimensional wake structure. In the present study, to investigate the structures of three-dimensional boundary layer separation, the flow of the separated region over a circular cylinder was observed by means of the flow visualization technique in the range of Reynolds numbers from 230 to 1340 in a towing water tank. As a result, it was determined that the three-dimensional features of boundary layer separation over a circular cylinder were characterized by the recirculating flow in the separated region, and the relationship between three-dimensional instability in the separated region and Reynolds number was investigated.

View full abstract

Experiments have been performed to investigate the effects of pressure gradient on the fluid flow and heat transfer in duct flow. Air flow in a fully developed turbulent state was decelerated through a linearly diverging section, and then accelerated through a linearly converging section. Two different ducts were tested, whose acceleration parameter K in the accelerating section was 2. 2×10_-6 and 4. 5×10_-6. The wall static pressure, the heat transfer coefficient and fluctuations of streamwise velocity as well as the time-averaged velocity profiles were measured. The heat transfer coefficient and turbulence intensity in the converging section were considerably lower than those in the diverging section, and the viscous sublayer expanded in the converging section. The behavior of the increase or decrease of the heat transfer coefficient was similar to that of the turbulence intensity in the near-wall region.

View full abstract

The problem of noise generation due to the interaction between flows and plate edges is treated analytically. In uniform flow containing vorticity waves, two semi-infinite flat plates are placed with the trailing edge of one plate and leading edge of the other being tandemly situated a finite distance apart. This flow is considered to be a simplified model for self-excited tones such as edge tone and cavity noise. An approximate solution to the sound pressure is obtained by the Wiener-Hopf technique, and the calculated acoustic field shows the characteristics of the trailing edge noise and the leading edge noise. The sound pressure level varies with peaks and troughs as the wave number increases, especially in the region upstream from both edges, and these peaks show a frequency dependence similar to edge tones. Such a selective response mechanism will be explained by the phase relationship between vortex and sound.

View full abstract

The oscillatory characteristics of a pseudo-shock wave with a Mach number of 1. 75 in a rectangular straight duct are experimentally investigated by static pressure measurements both on the duct axis and on the wall surface. The aspects of the pressure oscillation in the pseudo-shock region are analyzed from the treatment based on the intermittent characteristics, root-mean square, probability density function and cross correlation of the pressure signals. The correlation between the oscillatory characteristics and the flow mechanism of the pseudo-shock wave is also discussed in detail in the last section of the present paper.

View full abstract

A new technique of an active attenuator for pressure pulsation in liquid piping systems, which detects the progressive wave component of fluid-borne vibration and then controls the second source feedforwardly by this detected signal so as to completely eliminate the progressive wave in its downstream pipe by producing the canceling fluid vibration, has been presented, together with experimental results. As a preliminary investigation in this study, an experiment on the attenuation characteristics has been carried out for only the simple harmonic fluid vibration, though the present active attenuator using the transfer-function-based control method is also applicable to the wide-band random fluid vibration. A general-purpose digital signal processor has been used for the high-speed operation of converting the pressure difference signal produced by two pressure transducers into the progressive wave signal of fluid vibration and for the successive operation of producing the control signal to the second source (electrohydraulic servo cylinder). It was confirmed by the test in a model experimental pipeline that, when the present control system was operated, the pressure pulsation in a downstream pipe of the second source was reduced, as expected, to about 1/10 to 1/15 (that is, a transmission loss of about 20 to 25 decibels) of that in the case where the control system was not used over the frequency range of 20 to 250Hz.

View full abstract

Dependence of the flow behavior of underexpanded annular impinging jets on the nozzle contours is investigated experimentally. In the present study, five converging annular nozzles of different contours are fabricated and tested. The most important parameters of the annular nozzle contour are the ratio of the inner diameter d to the outer diameter D and the angle α of the ejection. The flow visualization using the shadowgraph method and the pressure measurements at the impinged surface show that the shock structure in the jets and the surface pressure strongly depend on the nozzle contour. It is demonstrated that the impinging jet issuing from an annular nozzle with large d/D and large α produces high surface pressure concentrated at the jet center. These results suggest that annular nozzles show great potential for assisting laser cutting.

View full abstract

The turbulent transition process in three-dimensional (3-D) bounday layers is known to be greatly different from and more complicated than that in two-dimensional (2-D) cases. As a result, the transition process in 3-D boundary layers is not well understood. When such a flow field is also critical to cavitation inception, which is often the case in existing fluid machineries, flow structures become further complicated. Transition phenomena might influence the cavitation inception and its behavior ; therefore, investigation of cavitation phenomena in relation to boundary-layer transition is essential in order to refine existing fluid machineries. However, in spite of such industrial importance, such problems have not yet been investigated. Cavitation first appears at the positions where pressure is minimum. Therefore, the vortices in boundary layers developing in wall surfaces of fluid machineries are possible candidates for the cavitation inception. In the transition regions of general3-D boundary layers, streamwise vortices often appear. It is interesting to investigate such vortices from the standpoint of cavitation phenomena. In the present investigation, the relationship between cavitation phenomena and boundary-layer transition on a 45-degree yawed cylinder is investigated using both wind and water tunnels. The yawed cylinder is an appropriate model for the purpose of such investigations while there appears cross-flow instability, and streamwise vortices will appear in the boundary layer. The transition process in the case of a yawed cylinder is often investigated in relation to a swept=wing boundary-layer transition study. Appearance of the cavitation streaks, which have a regularly aligned fine structure in the streamwise direction, on the cylinder surface show that cavitation selectively appears along each cross-flow vortex where a low-pressure zone exists in the transition region of the boundary layer. The wavelengths of the cross-flow vortices measured in a wind tunnel model and of cavitation streaks measured in a water tunnel model are in good agreement. Since the comparison of other physically measurable flow properties between the 3-D boundary-layer transition and cavitation phenomena came out reasonably well, it can be considered, in general, that the cavitation inception is enhanced by cross-flow vortices appearing in a 3-D boundary layer on a 45-degree yawed cylinder.

View full abstract

Among the problems in the application of high-water-content fluid, erosion induced by a cavitating jet is an important and complex one. A high-velocity jet accompanying cavitation can be observed in hydraulic valves and hydraulic pumps. This paper reports erosion induced by a cavitating jet flowing from a triangular notch. A pair of test pieces is used in each experiment. The first test piece has a triangular notch from which the fluid flows, forming a jet which impinges on the second test piece. A nondimensional parameter defined by the ratio of the eroded mass to the impinged fluid mass is used for expressing the strength of the erosion. The relationship between the eroded mass of each test piece and the mass of the expelled fluid is studied by statistical methods. The influences of the shape of the notch and the distance between the notch and the wall upon the erosion are also discussed.

View full abstract

Since a closed pipe maintains a low pressure near saturation vapor pressure during the formation of a cooling cavity, quickly opening a valve at an end of the pipe collapses the cavity due to the impressed atmospheric pressure, and it gives rise to a strong waterhammer of the pressure-impression type, in which the difference in pressures between the outside and inside of the pipe collapses the cavity. The above phenomenon is modeled by the rigid column theory. Formulas to estimate the maximum pressure rise are derived using the volumetric fraction of the cavity, the impression pressure difference, the mass density and the sonic velocity of the liquid in the pipe. Calculated curves are compared with the idealized experimental data. Because of air containment, the cavity does not completely vanish and some bubbles of air remain after the collapse. A weak buffer effect of the air in the cavity is shown quantitatively by the experiment.

View full abstract

A test sphere 100mm in diameter with programmable built-in processors was constructed to make surface pressure measurements during its free fall. The pressure data were stored in the body and transmitted to an external host computer after landing for further processing. The tests were performed under the smooth-surface condition and were compared with existing experimental data taken in wind tunnels, up to the Reynolds numbers of about 10⁵. The results showed that the gravitational acceleration was of little importance here, and that the apparatus worked with sufficient accuracy.

View full abstract

The effect of a tripping wire on a sphere in free fall has been investigated. The test sphere is of 1OOmm in diameter, and is equipped with built-in processors, such as a pressure sensor, a tension sensor, an amplifier, a memory, a power supply and a CPU. The tripping wire was formed by a ring of enptyv1mm wire attached to the surface at angles of θ_γ=25° and 55°, respectively, measured from the stagnation point. In the case of θ_γ=25°, the pressure distributions were nearly the same as those on a smooth sphere. In the case of θ_γ=55°, on the other hand, the effects of the tripping wire were striking, indicating a laminar-turbulent transition of the boundary layer at Re≒5×10⁴.

View full abstract

The present study proposes a new method of particle identification in three-dimensional particle-tracking velocimetry (3D PTV) using binary image data. It is based on the necessary condition for particle identification and the 3D cross-correlation method for evaluating the similarity of particle distribution patterns in two spaces. Computer simulations are carried out to investigate the effects of the sizes of particles, search subregion and identification subregion on the particle identification performance. The method has two key advantages. One is quick and accurate particle identification performance and the other is that the computer memory required for the computation of particle identification is generally small compared to other methods.

View full abstract

An alternative method for particle tracking, which is needed in PIV (particle imaging velocimetry), has been developed by applying the Kalman filter. The method improves the precision in measurements of particle positions and of their velocity components. Using this method, it becomes possible to track each particle during more than hundreds of time steps. These effects and the propriety of the applied Kalman filter estimator are proved experimentally by calibrating the behavior of particles which are fixed on a rotating disk. Moreover, the method enables us to acquire data from individual tracer particles in a three-dimensional unsteady flow field.

View full abstract

Effects of velocity gradient of linear shear flows at a high Reynolds number (Re, _s=4000-35000) on lift and drag applied to a spherical particle are investigated by measuring surface pressure distribution, using a pendant method, and by comparing computed and photographed trajectories of a sphere falling in upward shear flows. It is made clear that lift is applied to a sphere from the higher velocity side to the lower velocity side, that it increases as the velocity gradient becomes greater, and that drag is scarcely affected by the velocity gradient.

View full abstract

The present paper treats numerical simulations for detonation initiations behind reflected-shock waves in a shock tube. The two-dimensional, thin-layer Navier-Stokes equations with chemical effects are numerically solved by use of a method combining the Richitmyer-FCT scheme, the Crank-Nicolson scheme and a chemical calculation step. Effects of chemical reactions are estimated in the simulations by using a simplified reaction model in which rate coefficients are determined so that the induction times and the characteristic times of the exothermic reactions are fitted to data of a H₂/O₂/Ar mixture. Simulations are carried out, referring to experiments by several authors. Results of simulations reveal a mechanism of triple-shock generation in reaction shock waves. Computed flowfields for strong ignition in hydrogen and oxygen are in good qualitative agreement with visualized ones in the experiments. A simulation referring to mild ignition predicts such a feature that the ignition starts from distinct kernels. It is also predicted that an ignition occurs immediately behind a normal reflected-shock wave, but does not behind a bifurcated oblique shock wave.

View full abstract

The objective of this paper is to gain insight into the process of chemical reaction in turbulent mixing layers. To achieve this purpose, we have performed direct numerical simulations of three-dimensional chemically reacting mixing layers using a spectral method. The results are reported for the case of velocity fields with initial periodic excitation. From these results, we have reached the following conclusions : 1) on the streamwise vortical structures (ribs) in the braid region, the peak of the chemical production term exists ; and 2) the core region contributes significantly to the total production in the mixing layer, whereas local production rate is not so high.

View full abstract

The sputtering method is a widely used technique for fabricating thin films by use of glow discharge plasma. A Monte Carlo simulation method for calculating the growth rate of films is proposed in a general form. The method is applicable not only to cases where atomic collisions play a dominant role, but also to cases where the target suffers from nonuniform erosion. The method is used to find an optimum erosion distribution for a typical sputtering condition. First it is shown that films of uniform thickness cannot be achieved with a uniform erosion distribution. Next, the effect of nonuniform erosion on the thickness distribution is examined by changing the location of the maximum erosion rate. The overall growth rate of films is shown to decrease upon moving the location of the maximum erosion rate outwards. At a given pressure there is a corresponding optimum erosion distribution that yields a uniform film. The depth of the eroded dent has only a small effect on the thickness distribution of films, in particular, at high pressure.

View full abstract

A new FEM/FDM composite scheme is presented for two-dimensional incom-pressible flow problems combining the finite element method, which is best suited to analyze flow in any arbitrarily shaped flow geometry, with the finite difference method, which is advantageous in both computing time and computer storage. The combination of both methods enables large-scale viscous flow to be analyzed, which is crucial for both the detailed analysis of three-dimensional flows and the solving of problems of flow around moving bodies. A modified ABMAC method is used as the basic algorithm, to which a sophisticated time integration scheme proposed by the present authors has been applied. Numerical results including moving boundary problems indicate good agreement with experimental results.

View full abstract

Turbulent eddies in an ethylene jet flame are visualized by a laser-light sheet method. Tracer particles are soot spontaneously formed in the interface between fuel and air. The observation of the eddies in the flame shows that the scale of eddies in the flame increases with the kinematic viscosity so that the dissipation into smaller eddies is inhibited. It is also shown that large eddies in the horizontal cross section fold up the flame boundaries and make soot layers stretch thinner, thereby promoting air entrainment and fuel-air mixing. The eddies in the horizontal plane exhibit greater influence than the coherent ones in the shear layer. In addition, a hot and viscous soot layer formed in the outer region keeps the turbulent eddies from mixing with surrounding air, resulting in complicated, random patterns in the vicinity of the break point.

View full abstract

phase flow under a wide range of pressures from near atmospheric pressure to a highly elevated pressure of about 20 MPa. The liquid lumps of interest are liquid slugs, huge waves, disturbance waves and ephemeral large waves. Velocities were determined using time-series signals of cross-sectional mean liquid holdup which were electrically detected at two axially separated locations of the test tube. It is demonstrated how the mean values of liquid lump velocity and its standard deviation depend on the gas and liquid flow rates and pressure. It is obtained that the liquid lump velocity and its standard deviation are obviously affected by the pressure and are related to flow patterns. The comparison of liquid lump velocities between vertical upward flow and horizontal flow is also discussed.

View full abstract

Experiments have been performed to investigate melting heat-transfer characteristics along a horizontal heated cylinder immersed in a liquid-ice layer. A mixture of fine ice particles and ethylene glycol binary solution was utilized as the testing liquid ice. The experiments were carried out for a variety of parameters, such as heated-tube diameter, heat flux, and initial concentration of the aqueous binary solution. Attention was mainly focused on visual observation of the melting behavior and on its mechanism. The results revealed that the melting interface of the current liquid-ice layer might advance, in general, rather horizontally, unlike the normal melting observed in common rigid ice and PCM.

View full abstract

The heat transfer surface used in the experiments is a chrome-plated smooth brass pipe of 28mm diameter and 1400mm length. The film Reynolds numbers are varied in the range of 40-400, which includes the range of real machines. The average absorptions mass flux is compared with the numerical results derived from the laminar flow model proposed in the previous paper. The agreement is good up to 100 in Re_f, but the experimental values are far higher above 200. This may be attributed to the wave motions on the film surface. Pictures show that the film is completely smooth only at Re_f below 20. Beyond 20, ring-shaped waves appear, but the surfaces between the waves are relatively smooth at Re_f below 200. The measurements of the film thickness show that its minimum is constant at about 0. 38mm, but its maximum thickness increases for Re_f larger than 60.

View full abstract

The permeability is given by the hydraulic diameter and the product of the friction factor f_l and the Reynolds number Re. The former is obtained through the wire diameter and the porosity of screen wicks. The latter must be evaluated experimentally since it is influenced by the specifications of the screen and the wick thickness. Experiments have been conducted to investigate the effects of the characteristics of the wick on the permeability. The value of f_lRe decreases with the increase of the porosity, but still varies with the ratio of the wick thickness per layer to the wire diameter. A correlation between f_lRe and the ratio including the porosity is proposed. The permeability predicted by the correlation shows good agreement with the experimental results.

View full abstract

The temperature measurement accuracy of N₂ Q-branch CARS (Coherent Anti-Stokes Raman Spectroscopy) below 1000K has been improved by improving the spectral resolution of the detection system. The total spectral resolution was improved from 1. 3cm^-1 to 0. 6cm^-1 by dispersion magnification (×3. 6), which was effected by adding a magnifier between a monochromator and an intensified photodiode array detector. The accuracy of temperature measurement in the temperature range of 300-800K for pressures of 0. 1-3. 0MPa has been greatly improved from ±100K, in the case without the magnifier, to ±40K. The method reported in this paper has a potential advantage for practical use because of its convenience.

View full abstract

An exciplex-based fluorescence method was employed for remote, nonintrusive, instantaneous thermometry of a fuel droplet evaporating or burning in a stream of gaseous mixture. The fluorescence emission spectra from a droplet doped with naphthalene and N, N, N', N'-tetramethyl-p-phenylenediamine(TMPD)were measured with an optical multichannel analyzer. The effects of several parameters including the composition of the ambient gas and the concentrations of the dopants on the fluorescence emission spectra were studied. The results showed that the oxygen in the ambient gas was a major quencher for the fluorescence and that the fluorescence emission spectra varied with the concentration of naphthalene and TMPD. The velocity of the ambient gas had no appreciable effect on the fluorescence emission spectra. It was suggested that the temperature dependence of the ratio of fluorescence intensities at two different wavelengths for a droplet burning in air was the same as that for a droplet evaporating in nitrogen.

View full abstract

O₂ fluorescence induced by an ArF excimer laser was detected in a laminar diffusion flame under atmospheric pressure. O₂ concentration and temperature in the flame were estimated by O₂ fluorescence and Rayleigh scattering. The results obtained were as follows. (1) The O₂ fluorescence in the diffusion flame could be well predicted by the rate equations. (2) O₂ concentration in the flame was evaluated by O₂ fluorescence together with the temperature measured by Rayleigh scattering. Measured O₂ concentration agreed with that analyzed by a gas chromatograph.

View full abstract

The multisimulation model based on a so-called isothermal analysis has been newly developed for the primary performance prediction of both Stirling cycle machines and Vuilleumier cycle machines, which are realized as closed regenerative and external-combustion heat engines. The calculated results are compared with those of a so-called 3rd-order analysis, and the source of the difference between them is discussed. In addition, it is revealed by introducing an imaginary power piston into the present model that the Vuilleumier cycle machine is internally divided into a Stirling cycle engine and a Stirling cycle refrigerator which jointly own the imaginary power piston. This means that the input power into the internal Stirling cycle refrigerator and the output power from the Vuilleumier cycle machine which is obtained secondarily as a shaft power are produced by the output power from the internal Stirling cycle engine.

View full abstract