The secondary conservative finite difference for convective term is recognized as a useful tool for unsteady flow simulations. However, the secondary conservative convection scheme has not been extended to a moving grid. In this study, the secondary conservative convection scheme for ALE type moving grid simulations is proposed. For the moving grid simulations, the geometric conservation law (GCL) is known as the mathematical constraint on metrics and is interpreted as the condition for capturing a uniform flow. A new role of the GCL is revealed for the commutability and conservation properties of convective forms. The secondary conservative convection schemes for moving grids are then constructed for compressible and incompressible flows, respectively. In order to check the commutability of the convection schemes, a numerical test is done on a wavy grid problem. Then the reliability of the schemes is demonstrated on the piston problem and the flow around an oscillating square cylinder.

View full abstract

Experiments have been carried out to obtain the effective diffusivity of carbon dioxide for axial transport in the oscillatory flow through a straight pipe with circular cross-section for the case of a transition from laminar flow to turbulent flow. An experimental apparatus and procedures were devised to coincide with the theoretical situation. Results have been compared with the theoretical predictions by Watson and found to be in excellent agreement with both the theoretical and the experimental values in the case of laminar flow. The present results have less dispersion than those by the other researchers. In the case of turbulent flow, the results were larger than the theoretical values for the case of laminar flow because of the transition to turbulence. Using dynamically similar water, the axial velocity variations of the oscillatory flow were measured by the LDV system for several conditions. The turbulent intensity of the axial velocity and the effective diffusivity have been found to have a strong interrelation between each other.

View full abstract

In the present study, a transition mechanism to explain the change in the motion of a single ascending bubble from linear to zigzag was investigated experimentally. We focused on the relationship between bubble motion (gravity-center motion and surface motion) and hairpin-like vortex motion. First, the bubble motion was visualized using the infrared shadow technique (IST), and the gravity-center motion and surface motion were quantified. Second, the bubble wake structure and its motion were visualized with the laser induced fluorescence using HPTS (LIF/HPTS) method. From the LIF results, position fluctuations of the point of the hairpin-like vortex leg on the bubble rear surface (wake motion) were quantified. From these results, a close relation between the bubble motion and the wake motion was discussed. Furthermore, a relation between the wake motion and the lift force acting on the bubble was discussed. We concluded that the relations between them play an important role in zigzag motion. In addition, we tentatively concluded that bubble shape deformation observed just after the launch was the first trigger for the zigzag motion.

View full abstract

In this study, we developed a diaphragmless driver section with two pistons to generate the shock wave in small diameter tubes. An optical system using a He-Ne laser is constructed for the definition of a main piston opening time. The trajectories of main piston are measured under the several initial driver pressures. Additionally, the motion equation for the main piston is derived from a simple motion model. Moreover, the laser differential interferometer is constructed for the contactless measurement of the shock waves. The shock waves generated by our diaphragmless driver section, propagating in 2 and 3 mm inner diameter tubes are measured by the laser differential interferometer. The Mach number of the shock wave and the density ratio across the shock wave can be calculated by the interference signal obtained from the shock wave measurement. Additionally, the Mach number distributions along the axial direction of the tubes and the relation between the shock wave location and time (shock wave diagram) are obtained. Consequently, it is attributed to the fact that the friction effect between the test gas and the inner wall of the tubes becomes larger with decreasing the inner diameter of the tube.

View full abstract

An experimental study is performed for channel flow over a backward-facing step with the dielectric barrier discharge plasma actuator (DBD-PA). The plasma actuator consists of two metallic electrodes mounted on both sides of a dielectric plate. The actuator is installed on the corner edge of the step. The Reynolds number based on the mean velocity at the inlet and the step height ranges from 860 to 3240. The wall static pressure is measured, and the pressure loss coefficient is evaluated. The velocity profiles and turbulent intensities are measured by a PIV system at the Reynolds number of 1080, 1520 and 2170. The results show that the pressure drop at the step decreases by the effect of induced flow with the plasma actuator and the reattachment length is shorter compared with that of the steady flow condition. When the wave form with voltage of 6 kV and frequency of 2 kHz is applied to the electrode, the maximum pressure loss reduction rate of 31% is confirmed in the laminar flow region.

View full abstract

A high-speed train entering a tunnel generates a compression wave that propagates through the tunnel toward its exit. When the compression wave reaches the tunnel exit, a pressure pulse causing environmental problems (the micro-pressure wave) is radiated from the exit portal. The magnitude of the micro-pressure wave is approximately proportional to the maximum pressure gradient of the compression wave arriving at the tunnel exit. In a long Shinkansen tunnel with concrete slab tracks the compression wavefront steepens due to the nonlinear effect during its propagation because the whole surfaces of the tunnel being smooth. Hence, it is necessary to investigate the distortion of the compression wave in a tunnel and clarify the characteristics of compression wave propagation for estimating and reducing the micro-pressure wave. In this paper, we introduce a new simple equation governing the distortion of the compression wave propagating through a Shinkansen tunnel with concrete slab-tracks and propose a new simple scheme for a numerical calculation. A space evolution type equation with one variable is derived from the three conservation equations (mass, momentum, and energy including the wall friction and heat transfer terms) of the 1D CFD by assuming small disturbances excited by the compression wave. The numerical calculation scheme based on the simple equation remarkably reduces the computing time because its CFL condition is relaxed. The calculation results by the present scheme agree well with those by a conventional scheme based on the 1D CFD and the accuracy of the simple equation is verified.

View full abstract

Carbon capture and sequestration (CCS) is part of the global challenge of mitigating global warming and climate change. Geological sequestration of carbon dioxide (CO₂) is an immediately available and technologically feasible method for achieving a substantial reduction in carbon dioxide emissions into the atmosphere. Because the density of CO₂ is lower than that of water, injected CO₂ migrates upward in an aquifer owing to the buoyancy force. Therefore, a highly impermeable layer is generally employed to prevent CO₂ leakage from the storage reservoirs. For these reasons, assessment of the storage site, leakage risks, and storage costs are one of the main issues in CO₂ geological sequestration. The intent of the present study was to clarify the fundamental mechanism of CO₂ migration in a reservoir. The behaviour of liquid CO₂ in water-saturated silica packed bed was observed using high spatial resolution X-ray computed tomography. As the result, CO₂ behavior in the water saturated porous structure was visualized with 20 μm resolution, and it was found that the permeation and distribution of liquid CO₂ were strongly dominated by the local pore size and its structure.

View full abstract

Experimental investigations have been carried out on opposing mixed convective flows of water induced around a horizontal rectangular cylinder in the range of the Reynolds and modified Rayleigh numbers, Re_D =500-3000 and Ra_D^*=7x10⁷-10¹⁰. The flow and temperature fields around the cylinder were first visualized with dye and liquid crystal thermometry. The results showed that longitudinal vortices of which axes parallel to the flow direction appear over the upper surface of the cylinder when the buoyancy force beyond critical. The flow along the upper surface, then, separates at the edges of the cylinder. Thus, the side and bottom surfaces of the cylinder are covered by separation bubbles and wake. The local and overall heat transfer coefficients from the cylinder were subsequently measured. The coefficients from the upper and side surfaces showed marked increase with the buoyancy force. This yields higher overall Nusselt numbers than those of the forced convections. It was also found that the non-dimensional parameter (Gr_D^*/Re_D^2.5) well correlates the present overall Nusselt numbers.

View full abstract

Experimental investigations have been carried out on combined convective flows over upward-facing, circular heated disks placed in a vertical downward flow of forced convection. The experiments were performed with air in the ranges of the Reynolds and Grashof numbers as: 100 < Re_D < 2×10⁴, 10⁴ < Gr_D < 10⁷. The flow fields over disks are first visualized with smoke. The results showed that a twin-vortex appears over the disk intermittently when the buoyancy force is beyond a criterion, and that the onset of the above twin-vortex can be predicted with the Richardson number as Gr_D/Re_D² = 3.1. The overall heat transfer coefficients from the disks were subsequently measured. The results showed that the coefficients are increased significantly from those of the laminar stagnation point flows with the occurrence of the twin-vortex, and that the relative increase in the average Nusselt numbers of the combined convective flows against those of the stagnation point flows can be predicted with the above Richardson number.

View full abstract

Hot-wire anemometry (HWA) is used for measuring velocity fluctuations such as turbulent flows. It is generally operated in the following three modes; constant-temperature (CT), constant-current (CC) and constant-voltage (CV) ones. The constant-temperature anemometer (CTA) has long been the mainstream of normal turbulent flow measurement, while the others are rarely used in the measurement because of their insufficient response speed. However, since the constantcurrent anemometer (CCA) can be composed of simple circuits, the HWA will be realized at quite a low cost. In this study, the response characteristics of the CCA were theoretically analyzed, and a sophisticated compensation scheme based on the analysis has been proposed. The scheme is experimentally tested in a turbulent wake flow formed behind a cylinder. As a result, it has been confirmed that the root-mean-square (rms) velocities and the power-spectrum distributions compensated by the present scheme agree well with those measured with the CTA. Hence, the CCA provides reliable measurement of turbulent velocity fluctuations.

View full abstract

This study experimentally examined evaporation of the single droplet deposited on the heated surface where copper fine powders had been sintered. A sintered layer, which was made by Spark Plasma Sintering, ranges in porosity from 0.19 to 0.55, and in a thickness from 70 to 600 μm. Porous surfaces expanded the contact area of deposited droplet, and hence a contact angle was reduced as compared to non-porous surfaces. This area was also enlarged by an increase in pre-impact Weber number. A droplet being placed on the porous surface that had specific porosity, a wetted zone due to permeation instantly surrounded the droplet. Under a constant surface temperature, droplet lifetimes on the porous surfaces where the wetted zone did not appear shortened with increasing Weber number, since enlargement of the contact area and/or a droplet breakup occurs on the surface. For the surface where the zone largely formed, droplet lifetimes were almost independent of Weber number because of the larger evaporation interface that had grown by wicking. Mostly, the porous surfaces enhanced droplet evaporation in low temperature range.

View full abstract

Effect of non-uniformed electric field on PM emission from an acetylene diffusion flame was investigated experimentally. When a high voltage was applied to the flame, ionic wind blew from nozzle electrode toward ground electrode. The ionic wind increased with increasing the applied voltage. The velocity of ionic wind around the flame was measured by means of LDA (Laser Doppler Anemometer). Particle size distributions of PM emitted from two different types of flame were measured with SMPS (Scanning Mobility Particle Analyzer). As the applied voltage was increased, the flame length became short and fluctuation of the flame tip increased. In the case of “Open flame”, the PM emission more than 90% was suppressed by applying the electric field intensity of 150 kV/m. Especially, the emission of relatively large particles (30～500nm) decreased by the promotion of the mixing of fuel and surrounding air, while the emission of PM smaller than 20nm increased with increasing the applied voltage.

View full abstract

This paper discusses the operational characteristics of two emulsified blends of aqueous ethanol and diesel fuel in a DI diesel engine with supercharging and cooled EGR. Two emulsified blends: one with 25 vol% aqueous ethanol (E20W5, Ethanol: 20 vol%, Water: 5 vol%) and the other with 50 vol% aqueous ethanol (E40W10, Ethanol: 40 vol%, Water: 10 vol%) were examined. Silent, low NOx, and smokeless combustion was realized over a wide load range with the emulsified fuel, E40W10 when the quantity of pilot injection and the intake oxygen content were optimized. The results show that smokeless operation can be established with low pressure injection and low intake oxygen content. The degree of constant volume heat release and the indicated thermal efficiency with the emulsified fuels improved over a wide load range as the pilot injection quantity was increased and the heat release with the pilot injection was retarded to near TDC. The degree of constant volume heat release with the emulsified fuel also improved, mainly due to the reduction in afterburning with smaller quantities of the main injection.

View full abstract

In order to clarify the effect of heterogeneity of distributions of temperature and oxygen concentration in a combustion chamber on the soot and NO emissions from the diesel engine, the combustion characteristics and emissions of diesel spray flame achieved in a rapid compression and expansion machine (RCEM) at various patterns of temperature and oxygen concentration distributions in the chamber are investigated. The single-hole nozzle and multi-hole nozzle are used for the experiments. Results indicate that the NO and soot emissions from diesel combustion with single-hole nozzle are governed mainly by the amounts of oxygen and enthalpy entrained into the spray upstream the ignition region, when the temperature distribution and oxygen concentration distribution in a chamber are changed. In the case of multi-hole nozzle, NO emission is affected strongly by the amounts of oxygen and enthalpy entrained into the spray upstream the ignition region similarly to the results of the case of single-hole nozzle. However, the soot emission from the diesel flame with multi-hole nozzle is not affected by the heterogeneity of surrounding gas because the soot oxidation during late phase of combustion is governed mainly by the efficiency of utilization of air in a chamber which is affected significantly by the number of holes of the nozzle.

View full abstract

This experimental study was performed to investigate directly the local flame properties of turbulent propagating flames at the same weak turbulence condition, in order to clarify basically the influence of the addition of dilution gas to hydrogen mixtures on its local burning velocity. The mixtures having nearly the same laminar burning velocity with different dilution gas types and equivalence ratios were prepared. A two-dimensional sequential laser tomography technique was used to obtain the relationship between the flame shape and the flame displacement in a constant-volume vessel. The local flame displacement velocity S_F, curvature and stretch of turbulent flames were quantitatively measured as some of the key parameters of local flame properties on turbulent flames. Additionally, the Markstein number Ma was obtained from outwardly propagating spherical laminar flames, in order to examine the effect of positive stretch on burning velocity. It was found that the trends of the mean values of measured S_F with respect to dilution gas types and equivalence ratios corresponded well its turbulent burning velocity. The trend of the obtained Ma could explain the S_F of turbulent flames only qualitatively. The local burning velocity at the part of turbulent flames with positive stretch and curvature due to the Ma, S_Lt , attempted to be estimated quantitatively. As a result, a quantitative relationship between the estimated S_Lt and the S_F at positive stretch and curvature of turbulent flames could be observed only for mixtures with Le > 1 or Ma > 0, regardless of dilution gas types.

View full abstract

Experimental and numerical investigations of shock-compression ignition induced by a disturbance have been carried out. Experiments were performed in a square cross-section shock tube behind reflected shock waves at temperature between 1401 and 1681 K and the total densities were kept constant at around 30 mol/m³. The mixture used was a stoichiometric methane/oxygen/argon with 89.5 percent argon dilution. A thin plate or a fine wire was installed apart from the tube-end of the shock tube to generate density disturbance. The shape of shock-compression ignition flame under different disturbance was observed by direct and schlieren high-speed photography. The disturbance in the shape of a vortex appeared after the reflected shock wave passes through the obstacles. Ignition appears in the vicinity downstream of the obstacle in the tube-end side. The ignition flame is the same shape of the density disturbance. These disturbances made with obstacles promote ignition delay and fixes the position of ignition. The temperature in the disturbed region was evaluated using the temperature-dependent molar extinction coefficients of CO₂ at 220 nm. An infrared absorption method with an infrared He-Ne laser (3.39 μm) was used for measuring the density in the disturbed region. The second order explicit MacCormack-TVD scheme is used to solve two-dimension Navier-Stokes equations. The density of the downstream side of the obstacle is lower than that of the upstream side. Temperature in the disturbed region was higher than others; this high temperature was caused by reflected shock wave acceleration in a rarefaction wave region at the rear of the obstacle.

View full abstract

Effects of the swirl number S and the injection velocity ratio α on the structure of a rapidly mixed tubular flame have been experimentally investigated. Extinction limits, distributions of the flame chemi-luminescence, and distributions of local equivalence ratios of the unburned gas have been determined. Results show that, with the burners of different swirl numbers (S = 2.5, 5.0), the total equivalence ratio at the lean and the rich extinction limits are almost the same, however, both the limits shift on the relatively fuel rich side when α_st = 4.8, whereas the limits shift in the fuel lean side when α_st = 0.5 as compared with α_st = 1.0. According to the result of the flame chemi-luminescence measurement, combustion is intensified in relatively fuel rich side when α_st = 4.8, whereas combustion is intensified in the fuel lean side when α_st = 0.5. Results of the gas analysis have revealed that, these changes in the extinction limits and combustion intensities are attributed to the change in the local equivalence ratio Φ_local. At the center of the burner, a relatively fuel lean mixture is formed when α_st = 4.8, whereas a relatively fuel rich mixture is formed when α_st = 0.5 as compared with α_st = 1.0. These results indicate that, with use of the injection velocity ratio, a flexible, as well as an accurate structure control of rapidly mixed tubular flames is possible.

View full abstract

The viscosity of the endothelial cell membrane changes during cell migration or in response to shear stress exposure. Interestingly, the degree of viscosity change differs among the various regions of the endothelial cell. These local variations in membrane viscosity may contribute to spatially specific signal transductions. It is unclear, however, whether local membrane viscosity affects to membrane protein dynamics. To address this issue, we examined membrane protein diffusion at different sites in endothelial cells. We fused the photocromic fluorescent protein Dronpa to glycosil phosphatidylinositol-anchored protein (DGGPI) which diffuses on the extracellular surface of the endothelial cell membrane. The DGGPI diffusion coefficient D was 0.165 ± 0.013 μm²/s (mean ± SE) at 37°C. Examining different sites in the endothelial cells at 28°C (room temperature) revealed that the DGGPI diffusion coefficient at the cell edge was 27% lower than the value measured at the of cell body center.

View full abstract

[in Japanese]

View full abstract