Jet flow from an orifice nozzle is applicable to mixing and entraining of ambient fluid effectively because of a large shearing stress layer with a large velocity gradient at the jet edge due to the vena contracta effect, but the sudden contraction at the nozzle exit accounts for a large flow resistance. The centerline velocity increases in the x direction from the nozzle exit and reaches to the maximum of 1.2u_ce (where, u_ce is the nozzle exit maximum velocity) at x/d₀≈2 (where, d₀ is the nozzle exit diameter), this may be also effective to enhance the heat transfer performance of impinging jet. In order to reduce the flow resistance remaining a good mixing performance caused by the vena contract effect, use of a cone orifice nozzle has been considered. Moreover, to increase the turbulence for improvement of mixing properties the use of tab, lib, and notch for orifice nozzle is thought. In this study, the effects of use of a notch for the orifice nozzle on the flow characteristics are examined experimentally. Hot wire measurements were conducted to demonstrate the spreading or mixing performance of the notched orifice nozzle having a reduced flow resistance of the nozzle and an increasing turbulent intensity more than that for a conventional orifice nozzle.

View full abstract

The laminar-turbulent transition of a mixing layer induced by oscillating flat plates at an exit of a two-dimensional nozzle was experimentally investigated. The mixing layer was formed between the jet which issued from the nozzle and the surrounding quiescent fluid. The plates oscillated vertically in relation to the mean flow. The oscillation frequency was two orders of magnitude smaller than the fundamental frequency of the velocity fluctuation. Mean and fluctuating velocity components in the streamwise and normal directions were measured by hot-wire anemometers. The oscillation was found to be effective in enhancing the mixing, though the amplitude was the same order as the momentum thickness of the boundary layer at the nozzle exit. The disturbance traveled downstream as the convective instability, though it was damped only far downstream. The downstream development rate of fluctuating velocity in the normal component was larger than that in the streamwise one. Thus, the need for linear instability analysis of non-parallel flow was suggested. Streamwise variation of the fluctuating velocity and fluctuating energy production and convection rates, which contribute to the velocity, were examined. The streamwise variation in the streamwise component did not correspond to that of the normal component.

View full abstract

A sheathed nozzle with ventilation for abrasive suspension jets (ASJs) is developed as a means of extending the effective standoff distance in submerged environments. Submerged cutting tests are conducted with aluminum specimens at the jetting pressure of 30 MPa and the water depth of 100 mm. The nozzle used in the cutting tests is a conical convergent nozzle followed by a straight passage designated focusing section. The diameter of the nozzle and the inner diameter of the sheath are 1 mm and 3 mm, respectively. Air is sucked into the sheath via a plastic tube with an inner diameter of 6 mm. The air flow rate is measured by a float type area flow meter attached to the end of the tube. The abrasive is garnet having a mesh designation of #100. It is showed that the submerged cutting depth of the sheathed nozzle with ventilation is comparable to the cutting depth of the ordinary nozzle in air. The sheathed ASJ nozzle with ventilation is useful for extending the effective standoff distance between the nozzle and the work-piece.

View full abstract

The flapping motion in a two-dimensional turbulent jet is analyzed statistically. The analysis is done by applying a continuous wavelet transform with Gabor mother wavelet to the experimental result of hot-wire anemometer. The analysis results show that the flapping motion is clearly recognized by the present new method. The flapping motion occurs at random, and the average time interval of flapping motion is about 6.1 times the flapping periods (0.24 sec). The average duration time of the flapping motion is about 1.7 times the flapping period (0.07 sec) and the one that continues over 3.0 times the flapping period is very few. The negative value of the two-point spatial velocity correlation substantially depends on whether the flapping motion arises or not. From these results, it is found that the flapping motion in a two-dimensional turbulent jet has a very important role to the properties of the jet though the flapping motion is a rare phenomenon.

View full abstract

Impingement jet is widely used in industrial fields because it provides a high heat transfer coefficient near the stagnation region. However, few methods exist to control the impingement heat transfer. Recently, peculiar diffusion process “axis-switching” in three dimensional free jet has been reported, and the possibility of axis-switching controlling the diffusion and mixture process has begun to attract attention. In this report, we have studied experimentally the effect of the orifice configuration on the impingement heat transfer using non-circular orifices. In addition, “axis-switching phenomenon” was showed clearly by the flow visualization using the hydrogen bubble method. Orifice configurations are the regular polygons with 3 to 6 sides. Heat transfer experiments covered the distance between the orifice-to-target plate is 4 to 8 and Reynolds number is 5 × 10⁴ and the constant heat flux is 600 W/m². The flow was visualized in Reynolds number 1,500. In the free jet from the regular polygon orifice, when the number of sides on the orifice increased, the occurring location of "axis-switching phenomenon'' moved towards the orifice exit and the tendency of the iso-Nusselt number profile became the concentric profile upstream. With the decrease in the number of sides on the orifice, the iso-Nusselt number profile after “axis-switching phenomenon” was kept at the downstream.

View full abstract

This study is concerned with the characteristics of a radial jet flow which reattaches on a coaxially rotating disk that the disk sets parallel to a cylindrical nozzle. The velocity fields near the disk surface and in the jet were measured by a hot-wire probe to clarify the variation of the basic characteristics of the flow. As the offset distance from the wall to the nozzle (step height) becomes high, the reattachment position moves to the upstream side depending on the disk rotating speed. The basic characteristics of the jet until it reattaches on the disk are similar to the radial free jet. In contrast, after the reattachment, they show the change similar to the radial wall jet without step height. Furthermore, the velocity in the reattaching region fluctuates periodically, and the frequency of the fluctuation decreases with the step height and increases with the disk rotating speed.

View full abstract

Submerged water jet accompanied by ultrasonic-waves is considered to be able to warm up bones under the skin effectively and it is utilized at the rehabilitation facilities. However, there exist few reports about the submerged water jet with ultrasonic-waves. In this paper, an experimental study of ultrasonic- wave generated by a submerged water jet has been carried out using various size of the nozzle and varying jet velocity. Sound pressure level (SPL) is measured by a hydrophone in a water tank. It is found that the water jet without air mixed and that water mixed show higher SPL in the range of ultrasonic-wave than the conventional jet with air mixed, whereas SPL of the former at low frequency is decreased remarkably. Furthermore, SPL for the jet without air mixed persists to the downstream section. In the case of the jet without air mixed, a cap installed at the outlet of the nozzle is found to enhance SPL of the ultrasonic-wave, and optimum sizes of the nozzle are clarified. In addition, an increase in SPL in the range of the ultrasonic-wave is confirmed to be related to the cavitation, but the generation of ultrasonic-wave doesn't depend only on the cavitation.

View full abstract

Simultaneous measurements of instantaneous two-component velocities and concentration are conducted in an axisymmetric CO₂ turbulent jet diffusion field. Three probes are used. The first is the ‘double probe’ consisting of two concentration-sensitive I-type hot-wire probes for the simultaneous measurement of instantaneous axial velocity and concentration. The second is the ‘Chassaing type’ X hot-wire probe, which is insensitive to local concentration, for measurement of instantaneous two-component velocities in the CO₂ diffusion field. The third is the composite probe consisting of the double and Chassaing-type X probe for simultaneous measurement of instantaneous two-component velocities and concentration. The results show that instantaneous two-component velocities and concentration of CO₂ can be measured by using the present composite probe.

View full abstract

Effects of freestream turbulence on an axisymmetric CO₂ turbulent jet diffusion field is experimentally investigated. The composite probe consisting of two concentration-sensitive I-type hot-wires is used for the simultaneous measurement of instantaneous axial velocity and concentration. To measure grid turbulence without the jet, I- and X-type hot wire probes are used. The fractal square grid is used to generate freestream turbulence. The Taylor Reynolds number is Re_λ = 80 - 160. The results are compared with previous measurements in which weak freestream turbulence at Re_λ-35 is generated by the classical biplane grid (regular grid). For both flows, the mesh Reynolds number based on the grid mesh size M and freestream speed U₀ is 6,000 and the jet Reynolds number based on the diameter d (= 3 mm) of jet exit and relative velocity between freestream and exit velocity, U_J-U₀, is 5,000. The results show that profiles of normalized mean velocity and concentration in the CO₂ jet do not change at the present Re_λ in the freestream flow, while growth of half widths of mean velocity and mean concentration, turbulent intensity, turbulent mass flux and eddy diffusivity in the CO₂ jet are significantly increased under the effect of strong freestream turbulence. The eddy diffusivity in the CO₂ jet becomes approximately three times of that in the regular grid turbulence.

View full abstract

This report aims at investigating the relationship between preferred frequency and diffusion. The disturbance frequency for mixing control based on the preferred frequency occur in a jet flow initial area of a jet column mode is often selected to enhance the diffusion of the jet flow. When the diffusive mixing enhancement such as the gaseous fuel and air is assumed, it is general that a jet flow and a surrounding gas are different. So the jet flow, Air, CO₂, and He were used. And nozzles were three types, converging, orifice and pipe. First, the laser light sheet technique was applied to the visualization of a cross section of vortex ring. The distance until rolling up to the vortex ring goes away from the nozzle as increasing the density ratio. Next, the variation of velocity of a jet flow was measured by the hot wire anemometry and PIV to investigate the preferred frequency to determine the Strouhal number. The Strouhal number does not depend on Reynolds number and keep an almost constant value. These show the relationship between preferred frequency and diffusion by jet density and exit velocity distributions.

View full abstract

A strongly pulsed jet is considered as a method for rapid mixing of two different gases. Carbon dioxide gas with a pulsation was discharged from a convergent and round nozzle into still air. An oscillatory and reversed flow at the nozzle exit was produced by the pulsation with high intensity. The spatial and temporal variations of concentration and velocity were measured in the mixing region of the pulsed jet. The mixing rate of the gases is dramatically enhanced near the nozzle exit. Neither the mean concentration nor the mean velocity monotonically decays in the stream-wise direction, because the ambient air is drawn inside the nozzle periodically by the reversed flow. The direction of the velocity changes periodically at a certain boundary, which is the position where the mean concentration becomes the minimal value. When the mixing gas is aspirated inside the nozzle, the fluid lump of any concentration is separated to two fluid lumps because of the direction change in the velocity. The issuing mean velocity becomes larger than the velocity determined by flow rate of supplied carbon dioxide gas. The axial distribution of the mean velocity increases near the nozzle exit. There is also a phase difference between the concentration and velocity oscillations.

View full abstract

Simultaneous measurements on the velocity and concentration fields of matter in a turbulent round jet injected into a counter stream have been conducted by using a combined system of PIV and PLIF techniques. Experiments were carried out for four cases of jet to counter-flow velocity ratios: V_r = 2.4, 3.4, 5.5 and 7.8. Axial profiles of the concentration statistics along the jet centerline showed that both rms concentration fluctuation and streamwise turbulent flux of mass have two peaks in the near field of the jet and in the mean stagnant region. The radial flux was found to have an tendency to increase in an off-axis area outside the dividing streamline due to a circumferential wandering of the jet about the axis. These features for the axial and radial mass fluxes appeared more obviously as Vr decreased. By investigating the joint PDF between axial velocity and concentration, it was confirmed that alternate streamwise transport of dyed fluid in the jet and ambient fluid in the counter-flow is related with the enhancement in the streamwise turbulent flux of mass in the stagnant region.

View full abstract

In order to develop a new mixing procedure, we conduct DNS (direct numerical simulation) of vector controlled free jets. The inflow velocity of jet is periodically oscillated perpendicular to the jet axis. To realize the high accurate computation, the discretization in space is performed with hybrid scheme in which Fourier spectral and 6th order compact scheme are adopted. From view of instantaneous vortex structures, it is found that the flow pattern considerably changes according to the oscillating frequency, i.e., according to the increasing the frequency, wave like mode, bifurcation mode and flapping mode appear in turn. In contours of ensemble averaged streamwise velocity and turbulence kinetic energy, the jet diffuses largely in the oscillating direction. On the other hand, the jet width in the perpendicular to the oscillating direction is similar to that of uncontrolled jet. Further in order to quantify the mixing efficiency under the vector control, as the mixing measure, the statistical entropy is investigated. Compared to the uncontrolled jet, the mixing efficiency is improved in order of the wave- , the flapping- and the bifurcating mode. Thus the vector control can be expected for the improvement of mixing efficiency.

View full abstract

The formation process of a vortex ring generated by a pulsating jet flow, which is called the cyclic vortex ring, is investigated experimentally with PIV measurement. The waveform of the jet flow is conducted using sine curves by changing the amplitude of Reynolds' numbers Re₀ of 1200-5040 and Womersley numbers α of 12.5-24.4. The results indicate that the formation process of the cyclic vortex ring is different from the laminar vortex ring. Another vortex ring which is formed from separated wall boundary layers in the nozzle during the suction period, which is called the separation vortex ring, is transported toward the nozzle exit during the discharge period. This separation vortex ring interferes with the formation process of the cyclic vortex ring at the nozzle exit plane. The circulation of the separation vortex ring depends on the pulsating conditions of the jet flow. Distribution of the vorticity around the core center of the cyclic vortex ring is in good agreement with the Oseen vortex model. The formation process of the cyclic vortex ring is classified into three categories, depending on Strouhal number.

View full abstract

Basic experiments were performed to investigate the use of vortex rings as a means to transport a small amount of fluid. The compact and unchanging nature of vortex rings is suitable to obtain high-density distribution of the transported fluid within a specific confined region. A certain amount of fluid in a cylinder was discharged into a quiescent fluid in the water tank by a piston through an orifice 20.0 mm in diameter. The fluid distributions transported by a set of 10 vortex rings, by 20 vortex rings, and by a single, short puff-like jet were quantitatively analyzed and compared. For all ejection conditions, the total volume of discharged fluid and the average ejection velocity through the orifice are the same (78.5×10⁴ mm³ and 125 mm/s, respectively). The particle count method was used to quantitatively measure the distribution of the transported fluid. As a result, the vortex rings were found to be superior to the single short discharge for transporting a fluid to obtain a high density distribution in a narrow region.

View full abstract

The fluid flow and drag reduction of a circular cylinder with small grooves were investigated. The positions of upstream and downstream sides of the grooves on the cylinder, θ_f and θ_r, were 50 - 70 degrees, 55 - 74 degrees and 60 - 78 degrees from the stagnation point of the cylinder. Experiments were performed for the above three models and smooth cylinder varying the attack angle α = 0 to 45 degrees in the range of Reynolds number, Re = 10⁴-10⁵. The Strouhal number of the cylinder with grooves increases from 0.20 to 0.28-0.30, the base pressure coefficient rises from -1.4 to -0.8, then the drag coefficient decreases from 1.3 to 0.55 at α + θ_f < 80 degree, in the range of Re > 4×104. The surface oil-flow patterns show that the shear layers separated from the front of the groove reattach on the rear groove of the cylinder. This effect is due to the turbulent transition and turbulent separation. The wake width decreases and the location of the vortex formation region goes downstream.

View full abstract

An experimental study is performed for a turbulent channel flow obstructed with an airfoil type turbulence promoter. The Reynolds number ranges from 2.1×10⁴ to 8.4×10⁴, and the clearance between the wall and the promoter is changed in three steps. The total performance evaluation between heat transfer and pressure drop is estimated under the conditions of an equal pumping power. PIV is used for the measurements of velocity profiles, turbulent intensity and Reynolds shear stress. When the clearance is small, the heat transfer coefficient is augmented by the increase of turbulence intensity. The high performance is obtained by inserting the airfoil type turbulence promoter. At large clearance, the vortex shedding from the trailing edge of promoter is observed and high turbulence region is observed near the promoter. However, the excessive turbulence is rapidly decreased in the downstream direction, and the heat transfer coefficient becomes low value compared with that of the small clearance.

View full abstract

In this study, the flow features of vortex shedding from a pair of parallel arranged circular cylinders (same diameters and unequal diameters) oscillating along the direction of the flow were observed by visualizing water flow experiment at the ranges of the frequency ratio f/f_K=0-7, amplitude ratio 2a/d=0.125, 0.25, 0.5 and 0.75, gap ratio G/d=0.25, 0.75 and 1.75, and Reynolds numbers Re=470-670. The variations of mean vortex shedding frequency from each oscillating cylinder were investigated. As a result of the experiments, typical flow patterns of lock-in or un-lock-in state were shown every gap ratio G/d. It was found that the stage number of flow pattern was depended on the gap ratio G/d. In the cases of gap ratio G/d=0.25, there were four states, in the case of gap ratio G/d=0.75, there were six states, and in the case of gap ratio G/d=1.75, there were seven states. It was found that the measurement result of vortex shedding frequency depended on the observation point because the rearrangement of vortex formation had changed the characteristics of the flow. When the cylinders were oscillated to the direction of flow, the flow pattern of bias gap flow became not seen. The map of lock-in state in the lock-in range was obtained. It was found that the lock-in range and the lock-in form were more different from the case of single oscillating cylinder.

View full abstract

In this study, we clarified a formation process of coherent structures of turbulent flow around a three dimensional rectangular solid. The flow around the rectangular solid with the aspect ratio L/H of 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0 and 5.5, has been computed by the large eddy simulation at Reynolds number of 4 000. The computational results of the distribution of spectrum and Strouhal number correspond well with the flow around a two dimensional cylinder except for variation of peak value. And, in terms of Strouhal number, the result (L/H=1.0) corresponds with experimental result. In large aspect ratios, the visualization results show that a pair of hair pin vortex structures is shed from the vortex formation region alternately and periodically. The rectangular solid has two free ends. So, the aspects of the flow field around the three dimensional solid are different from that around the two dimensional cylinder, especially concerning the vortex formations. And, some characteristics of shear layer and vortex structures in a wake are found to be remarkably different with the aspect ratio. As for some vortices with characteristic aspect ratio, the formation and kinetic mechanisms of some vortices are described in details.

View full abstract

The present aim is to reveal the flow past a pipe which is immersed parallel to the mainstream at high Reynolds numbers. In a wind tunnel, we carry out (1) base-pressure measurements, (2) velocity-fluctuation measurements using a hot-wire anemometer and (3) flow visualisations by a smoke-wire method with PIV analyses, where we take consecutive picutures using a high-speed camcorder to obtain quantitative flow-field information such as velocity vector and vorticity. The tested parameter ranges are Re = 2.0×10³ - 1.3×10⁴, d/t = 4.0 - 10.0 and l/t = 1.0 - 10.0, where Re, d, t and l are the Reynolds number, mean diameter, thickness and length of the pipe, respectively. As a result, the Re effects are negligible. The base-suction coefficient -C_pb monotonously decreases with decreasing d/t, or with increasing l/t. We propose a unified formula to predict -C_pb, which are consistent with both a two-dimensional prism and a rod for l/t < 4.0 in addition to a ring. In contrast, the Strouhal number St almost coincides with that for a two-dimensional prism at any l/t, if we can detected any dominant frequencies. In addition, we conduct flow visualisations, and reveal the effects upon axisymmetry of wake. Finally, we classify the flow into three modes based on both periodicity and axisymmetry. Such a modal classification reveals that the enhancement of flow irregularity corresponds to the decrease of -C_pb.

View full abstract

In this study, we investigate flow control around a triangular prism in a uniform flow by utilizing receptivity to an external acoustic disturbance; we also investigate the mechanism of this flow control. The mean pressure fields on a prism surface are measured and the velocity fields around it are carefully investigated using Single type and X-type hot-wire probes. When a disturbance corresponding to the instability of a separated shear layer is added to a flow, the growth and the three-dimensionalization of the shear layer vortices related to flow randomization changes. Consequently, as the position of the Karman vortex shedding and the spanwise correlation of the vortex structure are changed, the force exerted by a fluid increases or decreases depending on the frequency of the disturbance.

View full abstract

This paper describes the analysis of flows around a straight wing vertical axis wind turbine. Hotwire measurements in two-dimensional wind tunnel and numerical simulations are conducted. The rotor has two blades, whose airfoil is NACA0012 and the chord length is 0.1 of the rotor diameter. Reynolds number based on the uniform velocity and rotor diameter is 8.0 × 10⁴. The tip speed ratio is 2.0. Incompressible Navier-Stokes equation is used for numerical simulation. A rotating coordinate system, which rotates at the same speed of the rotor, is employed. Both the experimental and numerical results are compared. As a result, the velocity fields show good agreements at each rotation angle, and the aerodynamic torque varies according to the flow condition, especially the flow separation.

View full abstract

Wind tunnel experiments are carried out to investigate interference effect of downstream strip plate of width w=d on the crossflow vibration of a square cylinder with a side length d=26mm. While sufficient suppression of Karman vortex induced vibration is not attained by plates with crossflow height less than l_d=4d, the galloping is suppressed even by a square plate, i.e. l_d=d, when the gap ratio 1<s/d<2.0. When a plate of l_d≤2d is set with a gap ratio s/d<1-1.4, a large vibration occurs at U>5m/s with the maximum amplitude at around s/d=0.3. The amplitude at s/d=0.3 increases divergently with flow velocity. Measurement of the lift force on the square cylinder and the velocity in the near wake for the corresponding fixed system shows no evidence for periodic change of flow responsible for the vibration. A prediction of this vibration based on the quasi-static hypothesis suggests that the vibration is caused by the fluid-elastic instability but not by a periodic vortex shedding. Hence, the name “Wake Body Interference Fluid Elastic Vibration (WBIFEV)” is proposed for this vibration.

View full abstract

Influence of the flow induced by dielectric barrier discharge (DBD) plasma actuators on the wing-tip vortex is investigated numerically and experimentally. The plasma actuators are installed on the suction side of the NACA0012 airfoil and operated in blowing and suction modes. For the numerical simulation, direct numerical simulation (DNS) based on the finite-difference immersed-boundary method is used. The DNS shows that the circulation parameter, which measures the strength of wing-tip vortex, is reduced by blowing as well as suction. At the same time, however, the lift-to-drag ratio is found to decrease by the actuation. In the experiment, a wing model with plasma actuators is set inside the wind tunnel and the velocity components are measured using a PIV system. Although suppression of wing-tip vortex is not confirmed due to insufficient strength of actuation and insufficient length of messurement area, the change of streamwise mean velocity profile is found to be similar to that of DNS.

View full abstract

The effects of voltage waveform characteristics on the flow induced by using a dielectric barrier discharge plasma actuator (DBD-PA) have been investigated in this study. It is known that the flow is induced in the direction from the upper electrode to the bottom electrode, then tangential one-way flow can be generated by using DBD-PA with asymmetric electrodes. We found, however, that it is possible to control the flow in certain conditions when offset voltages with a square wave are varied. Setting the negative offset, the flow is reversed from the bottom electrode to the upper electrode by applying square wave. It should be noted that it is not reversed by sine waves even with same amplitudes and frequencies. Our experimental results indicate that the voltage surge velocity is important to realize reversed flow induced by DBD-PA.

View full abstract

The objective of the present paper is to perform the effective control by the periodic disturbance in a backward facing step in low Reynolds number range. At first, in order to clarify the overall characteristics of separating shear layer, the flow behavior of the transverse vortices in the main flow and a reattachment length on a lower side wall were investigated by Hot-wire anemometer and Micro Flow Sensor(MFS). The characteristic frequency of the flow fluctuation was estimated to be about 36 Hz from FFT analysis. The change of the upstream vortex structure affects the downstream periodic flow behavior. To identify the periodic flow, the frequency dominating the separating shear layer was analyzed by the linear stability theory. From this result, the periodic disturbance was given to separating flow by means of the synthetic jets from the upper wall. The effective forcing frequency of the synthetic jets was evaluated from the reattachment length on the lower wall and the flow behavior in the separating shear layer.

View full abstract

The objective of the present study is to improve the design of a cylindrical water supply tank in order to achieve better displacement efficiency of old water in a cylindrical water supply tank. The present paper describes the results of experimental and numerical studies of the flow pattern and the displacement efficiency of old water in a cylindrical water supply tank. The displacement efficiency of old water in the tank having an inlet pipe with a diverging nozzle and an outlet pipe with a circular baffle plate is better than that of the other considered tank designs. The decrease in the displacement efficiency of old water in a tank that has the inlet and the outlet pipes at the center of the tank without a diverging nozzle and a baffle plate is related to the large fluctuation and the recirculation of flow in the tank. In addition, the results of the numerical simulation by FLUENT 6.3 for flow patterns and the mean residual concentration of old water in a cylindrical water supply tank were in good agreement with the experimental results.

View full abstract

The target of this research is to develop a micro electric power generator for a low cost small river monitoring device. First, the power generation by VIVs, its efficiency coefficient and the optimum condition of the generator were estimated from energy balance analysis based on the assumption that VIVs can be regarded as a resonance oscillation of a linear system. Second, water tunnel experiments were carried out and it was confirmed that the trailing vortex induced vibration (TVIV) occurs on a cruciform circular-cylinder/strip-plate system over a velocity range about 15 times wider than that of Karman vortex induced vibration (KVIV). Finally, power generation experiments were carried out by utilizing TVIV. The generator circuit consists of coils mounted on the circular cylinder vibrated by TVIV and magnets fixed on rigid supports. The generator is shown to extract energy from the water flow in the same way as a viscous damper over the expected velocity range. Although the efficiency coefficient of TVIV is lower than that of KVIV, it is more appropriate for natural rivers of which flow velocity changes greatly.

View full abstract

This study is concerned with the improvements of the Vortex in Cell (VIC) method for incompressible flow simulation. A discretization method employing a staggered grid is proposed to ensure the consistency between the discretized equations as well as to prevent the numerical oscillation of the solution. A method to modify the vorticity is presented to compute the vorticity field satisfying the solenoidal condition. A single-stage calculation method for the convection of vortex element is also proposed to reduce the computational time. To demonstrate the validity and applicability of these methods, the flows in a cubic cavity are simulated by the VIC method. The simulation demonstrates that the solenoidal condition for the vorticity is satisfied and that the velocity fields are in good agreement with the existing results. The Taylor-Görtler-Like vortices are successfully captured at Re=3200. It is also confirmed that the calculation for the convection of vortex element requires less computational time than the 2-stage Runge-Kutta method.

View full abstract

Large eddy simulation (LES) was performed for a stably stratified turbulent flow in a wavy wall channel. The numerical results at four grid resolutions were compared with existing data of experiments and direct numerical simulations, and revealed that the grid dependency of a subgrid-scale model in the stably stratified flow is quite larger than that in neutral flow. Flow becoming stably stratified, the flow pattern and the instantaneous turbulent structures near the wavy surface do not markedly change, whereas the turbulence intensity considerably decreases between the wavy surface and the center of the channel. The budgets of turbulent energy and Reynolds stresses suggest that energy transports between Reynolds normal stresses are activated far from the wavy surface, and spatial transports of turbulence are suppressed near the surface under stable stratification.

View full abstract

The effects of grid-generated turbulence on a turbulent boundary layer are investigated by means of a direct numerical simulation (DNS). A square turbulence-generating grid, on which the velocity components are set to zero, is numerially constructed near the entrance to a rectangular parallelepiped domain. The nonslip condition is applied to the lower boundary to develop a turbulent boundary layer downstream of the turbulence-generating grid. The fractional step method is used to solve the governing equations. The results show that streamwise and vertical turbulence intensities and Reynolds stress normalized by the inner parameters are suppressed in the log-law region by the grid turbulence.

View full abstract

The turbulent flowfield of turbulent jet issuing from rectangular nozzle (Aspect Ratio=12.5) with a rectangular notch at the midspan, has been investigated experimentally. Four aspect ratios of rectangular notch (NAR) used in this experiment were 2.5, 7.5, 12.5 and 165. The Reynolds number based on the nozzle width d and the exit mean velocity U_e, was kept constant 30000 (NAR=2.5 and 7.5), 15000 (NAR=12.5) and 13000 (NAR=165), respectively. From this experiment, it was revealed that the attachment of the rectangular notch to the rectangular jet restrained the increase of the turbulent velocity scales in the cases of NAR≥7.5. The principal strain rate S_xy on the y axis in the cases of NAR≤2.5 indicated the direct contribution to the production of the turbulent kinetic energy near the jet center in the upstream region, however, the development of the turbulent velocity in the cases of 7.5≤NAR≤12.5 was suppressed because of S_xy=0. Furthermore, the extra strain rate S_zz caused by the secondary flow on the z axis contributed to the production of the turbulent kinetic energy near the jet center in the upstream region, and the other extra strain rate S_yy on the y axis contributed to the suppression of the turbulent velocity.

View full abstract

Surface pressure distributions on a cone in a thin boundary layer have been measured to clarify the relation between the distributions and the flow structure formed around the cone. The apex angle 2α of cone is in the range of 15 to 90°and the height is 50mm. The Reynolds number based on the height is 2.31×10⁴ , and the relative boundary layer thickness is 0.44. The main directions of limited streamlines on the generatrix at θ=0°are predicted from pressure gradient forces in height and circumferential directions, and the front stagnation point which the streamline faced on the cone surface bifurcates downward and circumferentially on the generatrix will be determined as the position where the ratio of the pressure gradients forces is order of 1. Then, the height of the point have a peak value around 2α=45°. The maximum pressure coefficient at θ=0°can be expressed as the cosine law with respect to α . The surface pressure above the position of the maximum pressure at θ=0°can be related to 2α as well as the angle which the curved streamline in upstream region faces on the cone surface. The pressure drag coefficient of the cone mainly depends on the surface pressure distribution at θ=0°varied with 2α .

View full abstract

Experiments have been performed to investigate the effect of an adverse pressure gradient to the scaling law of a turbulent boundary layer. An approximate equilibrium boundary layer was achieved using a power law distribution of a free stream velocity with an exponent of -0.188. The wall shear stress was measured with a direct measurement device equipped with a zero displacement floating element mechanism. The scaling law was examined by the use of the friction velocity and a local velocity scale estimated from the shear stress profile. The later scale is derived from the integration of the boundary layer equations in conjunction with the order analysis in the limit of that pressure gradient parameter tends to zero and can be expressed as a function of friction parameter, pressure gradient parameter and height from the wall. The mean velocity gradient normalized with distance from the wall and the friction velocity depends on the height, but with the local velocity scale has a constant value in the layer where the PDF profiles of streamwise and normal fluctuating velocities are similar. The inverse of the constant value that is Kármán constant is 0.45. The modified logarithmic velocity profile is proposed from the local velocity scale and Kármán constant of 0.45 and is well consistent with the present experimental data.

View full abstract

Flow characteristics and fluid force on a sphere rotating along with axis perpendicular to mean air flow were investigated using Large Eddy Simulation (LES) at two different Reynolds numbers Re_p of 1.0×10⁴ and 2.0×10⁵. As a result of simulation, opposite flow characteristics around the sphere and displacement of the separation point were visualized depending on the Reynolds number even though rotation speed according to the Reynolds number is the same. When the sphere rotates at some specific rotation velocity and at Re_p = 1.0×10⁴, flow characteristics agree with the flow field explained in the Magnus effect. While sphere rotates at the same rotation velocity while increasing Re_p to 2.0×10⁵, separation point moves in opposite direction and wake appears in the different direction. The reason of the negative Magnus force was discussed in terms of the boundary layer transition on the surface.

View full abstract

The present study clarifies the effects of dimples for drag and lift on a sphere with rotation. The sphere with dimples has 328 arc type dimples on its surface uniformly. The present study measures velocity distribution in wake and pressure distribution on a sphere surface, and visualizes flow around a sphere by the spark tracing method for the case in which Reynolds number Re=0.4×10⁵-1.3×10⁵. For the case of a smooth sphere, separation point on deceleration side shifts downstream remarkably as spin rate ratio (circumferential velocity of a sphere / free stream velocity) increases. The pressure distribution on deceleration side decreases compared to that on acceleration side. The wake becomes small and shifts opposite direction of rotation. Therefore, the drag reduces, and the lift becomes negative. For the case of a sphere with dimples, this phenomenon occurs for the case in which critical Reynolds number Re=0.5×10⁵. The separation point on acceleration side shifts downstream, and the separation point on deceleration side shifts upstream as spin rate ratio increases for the case in which super critical Reynolds number Re=1.3×10⁵. The pressure distribution on acceleration side decreases compared to that on deceleration side. In addition, the pressure distribution on outside dimple decreases compared to that on inside dimple. The wake becomes large and shifts direction of rotation. Therefore, the drag and positive lift increase.

View full abstract

In this paper, extended immersed boundary method (IBM) for liquid-solid two-phase flow with heat transfer is proposed. The velocity and pressure field are solved based on IBM of body force type. To treat heat transfer at the interphase, the temperature field is computed with the thermal conductivity modelled by considering the direction of interface and heat flux. Present heat conduction model is validated by heat conduction problem of compound cylinder. In addition, the numerical stability is improved by the predictor-corrector method. It enables the simulation of flow laden with solid particles of neutral density. Then, the applicability of present method is shown by direct numerical simulation of two-dimensional cavity flow including multiple particles, and the influences of particle size, layout and behavior for heat transfer properties are investigated.

View full abstract

Instability characteristics generated with partial flow rate in a mixed-flow pump have bad influence during the pump operation. In this paper, a method to control blade loading distributions of the mixed-flow pump by a simple and easy method is shown. This method was applied to control of the blade loading distribution of the blade tip in mixed-flow pumps. Furthermore, performance and the internal flow of these pumps were examined experimentally. As a result, head and pump efficiency fell at flow rate of the best efficiency in the impeller of aft-loading, but the instability characteristics were largely improved. In addition, a reverse flow suddenly developed at the impeller inlet tip side with the decrease of the flow rate in the impeller of fore-loading, but it was shown clearly that the reverse flow did not suddenly develop in the impeller of aft-loading.

View full abstract

The purpose of this research is to elucidate the mechanism of aerodynamic force generation and control in flapping flight of a mosquito. In this paper, a tenfold enlarged realistic model of the flapping wing was used to simulate the motion of the mosquito's wing and flow pattern around it. Experiments were conducted using a very low speed wind tunnel to coincide the condition of Reynolds and Strouhal numbers with those of real mosquito. The flow field around the wing model was analyzed by a stereo-PIV system. The aerodynamic forces generated by the model wings were calculated by using a momentum conservation law applied to the flow field. In these experiments, two unsteady ring vortexes around model wings were observed. The magnitude and direction of the jet flow induced by the ring vortex were greatly influenced by the attack angle. It was found that the lift was generated during the down-stroke and the thrust during the up-stroke in the real mosquito's flight.

View full abstract

In this study, a generating process of microscale local countercurrent flows between parallel electrodes with an applied electric field was investigated using a numerical simulation model for the computational analysis and experimental measurements. An insulating fluid containing dispersed dielectric particles exhibits rotational flow pattern generated with local countercurrent flows under an alternating electric field. These flow structures are formed under high-electric-field intensity and low-frequency conditions, and are generated after the electric field inversion. This phenomenon has enormous advantages for manufacturing processes and biomedical applications. A numerical model of this phenomenon is important to investigate the generating mechanism and to develop mechanical and biomedical applications. In this study, the Eulerian-Lagrangian method, which is both an analysis of forced particle dynamics in an electric field and a computational fluid dynamics (CFD) analysis, was used as a numerical simulation model. In this model, the forces acting on the polarized particles and local dielectric forces were calculated based on the local nonuniform electric field. Furthermore, the surface property of an individual particle was designed as the microscale diamond particle of surface termination. As a result, the calculated flow structures agree with the experimental results of the generating process of the rotational flow. It is clear that this numerical simulation model is able to calculate the generating process of the rotational flow under an alternating electric field.

View full abstract

Creep tests, flow birefringence measurements, and flow visualization experiments in shear flows between parallel plates were carried out for aqueous solutions of cetyltrimethylammonium bromide (CTAB) and sodium salicylate (NaSal) with excess of NaSal. The mol concentration of CTAB was fixed to 0.030 mol/l and the concentration ratio of NaSal to CTAB was varied. The creep tests indicate that it requires long time to develop micellar network structures for low to moderate concentrations at low shear stresses. Under conditions of shear rate at which the shear-thickening property in viscosity is observed in a flow curve, periodical changes in flow birefringence and shear stress occur and these frequencies are almost the same as that of the emergency of white turbidity observed in the flow visualization. These results indicate that relatively large structures of micellar network are induced by the shear flow and repeatedly appear and disappear.

View full abstract

The present study experimentally investigated the performance and emission characteristics of the diesel engine with small amount of hydrogen added to the intake air at high EGR rate. The EGR rate, the diesel-fuel injection timing, and the hydrogen fraction in the intake mixture were varied while the available heat produced by diesel fuel and hydrogen per second was kept constant under high load condition in naturally-aspirated diesel engines. In the case of the diesel-fuel injection timing of 2 deg.BTDC with 3.9 vol% hydrogen addition, the smoke emission value was almost 0 %, NO emission was low, the cyclic variation was low, and the maximum rate of incylinder pressure rise was acceptable under a nearly stoichiometric condition without sacrificing indicated thermal efficiency.

View full abstract

In this paper, an Electrostatic-Cyclone DPF (Diesel Particulate Filer) has been developed to reduce the Particulate Matter (PM) from a marine diesel engine. Size distribution of the PM in the exhaust gas of four-stroke marine engine (1200rpm, 67.3kw) was measured by a Scanning Mobility Particle Sizer (SMPS) and a Low Pressure Impactor (LPI). The experimental results demonstrate that particles of less than 1 μm in diameter account for a major portion of the PM and that the DPF can collect fine particles in the PM (not exceed 500nm in diameter) from the engine exhaust gas. The DPF shows high collection efficiency for particles under 200 nm in diameter.

View full abstract

The gas turbine system known as AHAT (Advanced Humid Air Turbine) has been drawing attention. This system employs a humidification tower to increase the air mass flow at the compressor outlet. To estimate the humidifying transients, a dynamic model of the humidification tower was developed. This paper describes the humidification tower dynamics based on simultaneous differential equations considering heat and mass equilibrium of humid air, feed water, vessel materials, and packing materials. Values calculated using the model were compared to the experimental values obtained from laboratory-scale experiments. The calculation error margin was 3.8 degrees for air outlet temperature and 10.9% for air outlet absolute humidity. The developed model was then applied to a demonstration-scale conditions. The maximum change rate of absolute humidity at the humidification tower exit for startup and shutdown operations was 8.6%/s for 30 seconds. The maximum change rate of absolute humidity was influenced by the minimum area flow rate which corresponded to performance of packing materials.

View full abstract

A novel measurement technique to visualize the spatial temperature distribution of a non-isothermal flow is developed. Two-dimensional temperature profiles are obtained by quantitatively tracking the path of a rod-type probe with a high-speed CCD camera, and superimposing it on the image of a measured object photographed by the same camera. The rod-type probe consists of 8 two-thermocouple sensors set in line. Each two-thermocouple sensor used in the present experiment is composed of two fine-wire thermocouples (type K) of unequal diameters, i.e., 51 μm and 76 μm, respectively. The key technique is an adaptive response-compensation scheme by which the response lag of the temperature sensors can be adequately compensated; hence the temperature profile can be reconstructed accurately. It is demonstrated that the non-isothermal field can be visualized readily and quantitatively.

View full abstract

This paper describes physics-based moisture loss models for designing wet-steam turbines. Steam turbines in wet-steam conditions experience thermodynamic and mechanical losses caused by the presence of moisture. The developed moisture loss models are composed of six categories : supersaturation loss, condensation loss, accelerating loss, braking loss, capturing loss and pumping loss. These losses are calculated according to the wetness levels in the nozzle & bucket rows, and defined by the functions of steam properties and design factors of turbine stages. The supersaturation loss produced by the non-equilibrium expansion of the wet steam is modeled using the pressure ratio and the isentropic exponent in the supersaturated steam. The loss models are applied to the full-size test turbine data for the prediction of the overall moisture losses. As a result, the developed loss models are found to be very effective for estimating the moisture losses of turbine stages.

View full abstract

The n-heptane and toluene blended fuel exhibits the dual phase high temperature heat release (DP-HTHR) HCCI combustion and it has the potential to achieve the high load HCCI operation. Three topics were investigated in this paper. The first topic was the possibility of DP-HTHR combustion with the blends of the refinery-based gasolines. Three prototype fuels (GAS2, GAS3, and GAS4 fuels) were evaluated. The second topic was the relation between HCCI knocking and heat release phasing. The combustion parameters, which determined the maximum pressure rise rate, were statistically investigated. Finally, the combustion images of DP-HTHR were recorded with a high speed camera, and the combustion process of DP-HTHR was investigated by the video images and heat release analysis data.

View full abstract

Melting of an ice cylinder inside a cylindrical cavity including a calcium chloride aqueous solution is considered experimentally. The ice cylinder rotates with prescribed speeds and cylindrical cavity inclines with prescribed angles. When the inclined angle is small, melting mass of ice cylinder is enhanced by rotation, and melting front shows a specific configuration similar to a “screw”. On the other hand, when the inclined angle is large, melting mass of ice cylinder is reduced by rotation, and melting front shows smooth in spite of rotation of the ice cylinder. We infer that ice cylinder and melted water rotate together due to a viscous force in the case of large inclined angle. Therefore concentration gradient near the melting front becomes small, melting mass is reduced and melting front becomes smooth.

View full abstract