Blood pumps for the cardiopulmonary support system with ECMO (extracorporeal membrane oxygenation) need a higher pump head. Existing single stage centrifugal pumps can generate the higher pump head with higher rotational speed of the impeller. However, the large shear stress due to the higher rotational speed can occur and lead to hemolysis. In order to realize a blood pump with higher pump head, higher anti-hemolysis and thrombosis performances, the basic study on the development of unprecedented multistage blood pump was conducted. In consideration of the application of the blood pump for pediatric patients, miniature two-stage centrifugal blood pump with the impeller's diameter of 40mm was designed and the performance was examined in experiments and computations. Some improvements for a multistage blood pump with higher anti-hemolysis and thrombosis performances were suggested.
For a cardiopulmonary support system with a membrane oxygenation such as a percutaneous cardiopulmonary support (PCPS) or an extracorporeal membrane oxygenation (ECMO), a blood pump needs to generate the pressure rise of approximately 200mmHg or higher, due to the high hydraulic resistances of the membrane oxygenation and of the cannula tubing. Using existent single stage rotary pumps with higher pressure rise result in the higher rotational speed of the pumps and possibly lead to the higher amount of hemolysis due to the larger shear stress acting on blood cells. In order to realize a blood pump with higher pressure rise, higher anti-hemolysis and thrombosis performances, the development of novel centrifugal blood pump consisting of two-stage has been conducted by the authors. As the conditions when the hemolysis and thrombosis occur are not clear and the type number of blood pumps is quite small, the design method of blood pumps has not been established yet. In the present paper, effective attempts to decrease the wall shear stress and to suppress the stagnation are introduced for the prevention of hemolysis and thrombosis in blood pumps. The hemolysis test was also carried out and it was clarified that the decrease of wall shear stress is effective as a guideline of design of blood pumps for improving the anti-hemolysis performance.
Numerical simulation code for compressible flow based on OpenFOAM-1.6 was developed to compute a high Mach number flow. Total enthalpy transport equation was solved to calculate temperature and a flow around a rotor was solved in whirling rotating frame to predict a hydraulic force on an eccentric whirling rotor. Straight and high-low labyrinth seals were simulated by the developed code and numerical results regarding hydraulic force were compared with experimental ones to validate the prediction accuracy. The non-diagonal component of spring constant was predicted well, but the diagonal component of damping constant was largely under-predicted. The maximum and minimum pressures were observed at the circumferential positions of 270° and 90° respectively in inlet region of the straight labyrinth seal. Therefore large hydraulic force in the whirling direction was generated. On the other hand, the maximum and minimum pressures were observed at 180° and 0° respectively in the high-low labyrinth seal. Therefore the hydraulic force in the whirling direction was lower than that of the straight labyrinth seal. Pressure distribution in the circumferential direction was uniform in a large clearance region surrounded by two fins in the straight labyrinth seal, where the hydraulic force in whirling direction was largely decreased.
An undershot-type cross-flow water turbine does not require a casing for open channels and hence considerable simplification is attained. However, because this water turbine does not have a casing and has a free surface, it is believed that the internal flow considerably changes with the rotational speed. In this study, the performance and internal flow of the undershot-type cross-flow water turbine at a constant flow rate were investigated by an experiment and numerical analysis, and the following facts were clarified. The first-stage inlet area and the second-stage inlet and outlet areas are reduced with increase in rotational speed. Shock loss at the second- and first-stage areas is dominant in low- and high-rotational-speed regions, respectively. Moreover, shock loss has a considerable influence on the performance of this water turbine in low- and high-rotational-speed regions.
The calibration facility was newly constructed for high-pressure hydrogen gas flow meters. This facility can supply hydrogen gas of 350 Nm3/h at 35 MPa continuously for about 10 min. The calibration facility has the multi-nozzle calibrator, in which five critical nozzles calibrated by the NMIJ standards, are equipped, and can measure hydrogen gas flow rate within the relative standard uncertainty of 0.18 %. Also, the Coriolis flow meter for high pressure hydrogen were calibrated and evaluated by this calibration facility, and its output was scattered within about ±1.5 %. The present results show that the calibration facility constructed is useful for the evaluation of the high-pressure hydrogen gas flow meters, as well as the calibration.
A new probe for the simultaneous measurement of three velocity components and static pressure in turbulent flows is developed and its measurement accuracy is checked. The probe consists of two X-type hot-wire probes and a static pressure probe that is placed at the center of hot-wires. The static pressure tube is manufactured by using a micro fabricated metal tube and drill to improve the spatial resolution. The external diameter of the static pressure tube is 0.3 mm and its internal diameter is 0.2 mm with 8 static holes on the wall of the tube. The measurement error in the pressure measurement caused by the cross-flow is compensated by the measured instantaneous three velocity components by two X-type hot-wire probes. The measurement results show that the cross-streamwise profile of the mean velocity, RMS value of the fluctuating three velocity components, and RMS value of the fluctuating pressure in a plane jet agree with those obtained in the previous studies. Further, it is also found that the profile of the production term and diffusion term in the turbulent energy transport equation which is directly estimated by the measured data is more improved than our previous studies. This is due to the improvement of the spatial resolution of the combined probe and the measurement accuracy of the velocity-kinematic energy correlation by the measurement of all three velocity components.
We attempted to generate homogeneous turbulence of high intensity and large scale using a multi-fan wind tunnel. Input signals composed of multiple sinusoidal waves of constant amplitude were fed to each fan with quasi-random phases (“random-phase method”). This method produced approximately homogeneous and isotropic turbulence within a short distance. Although a pulsating component survived for the random-phase method using a single sinusoidal wave, it disappeared when multiple sinusoidal waves were applied. The initially provided rectangular spectrum rapidly changed its form as turbulence convected downstream, and asymptoted to a well-known broader spectrum with a few decades of inertial subrange of slope close to -5/3. The traditional method of generating approximately isotropic turbulence is by means of a grid, for which turbulence Reynolds number Reλ is low, typically in the range 50 to 150. The present method achieved a turbulence-Reynolds number of 753 for the largest amplitude of the input signals. The turbulence characteristics were compared with those of previous studies using active tubulence generation, such as a Makita-style active grid.
A drop actuation utilizing the wettability change by laser beam irradiation was studied by numerical simulations. The drop is actuated because surface tension force acts on the irradiated side more strongly in the wall parallel direction due to the smaller contact angle on that side. On the other hand, the temperature gradient due to laser irradiation causes surface tension gradient, then so-called Marangoni flow is caused. Each flow direction due to the wettability effect and the Marangoni effect is opposite. Those effects were evaluated by numerical simulations by a front-tracking method, which can properly evaluate the surface tension effect even on the solid surface and even with spatial gradient. As the results, we found that a simple model considering the macroscopic force balance could predict the moving condition due to wettability change fairly well although it could not predict the condition of movement caused by Marangoni effect. It is shown that the present numerical approach is very useful to predict such complex interfacial phenomena.
An increase of turbine blade loading reduces the numbers of blade and stage. As a result, the downsizing and the weight reduction of gas turbines can be achieved. On the other hand, an increase of blade loading makes the secondary flow much stronger because of the steep pressure gradient across the blade-to-blade passage and deteriorates the turbine efficiency. In this study, the computations were performed for the flow in an ultra-highly loaded turbine cascade with high turning angle in order to clarify the effects of the incidence angle on the secondary flow and the loss generation. For the verification of the computed results, the experimental oil flow visualizations were also conducted on the blade surface and the endwall. Moreover, the measurements of blade surface static pressure were performed at midspan of the blade. The computed results showed good agreement with the experimental results. The influences of the incidence angle on the secondary flow structure, the total pressure loss, the secondary flow kinetic energy and the blade loading distribution were examined in detail. The positive incidences not only strengthened the horseshoe vortex and the passage vortex but also induced the newly formed vortex on the endwall, which was generated by the interaction of the reversed flow along the suction surface caused by the strong passage cross flow with the incoming boundary layer. Moreover, the newly formed vortex influenced the formation of the pressure side leg of horseshoe vortex.
Microspray mode of electrostatic inkjet has been examined experimentally for the application of precision film coating. One of the critical issues lies in productivity because the amount of liquid jetted from a single nozzle is too small to obtain sufficient coating speed. To overcome the situation, a possibility of jetting two types of functional liquids from multi-nozzle was investigated. Both end-nozzles were used as electrodes to control jetting direction. According to the increment of applied voltage, jetting mode was varied from dripping to stable cone-jet. Although the evolution of jetting mode was qualitatively the same as that of a single nozzle, the much higher voltage was required for the multi-nozzle to obtain the stable jetting mode. Calculated results suggested that the higher voltage adjusted the force from electric field at the tip of the cone jet. From coating experiments on a rotating drum, it was demonstrated that the multi-nozzle with control electrodes at high applied voltage could jet well-oriented fine droplets to acquire thin or thick flat film. It was confirmed that the coating speed was improved according to the number of the nozzles.
A new discretization scheme for a Cartesian grid method is proposed. The Navier-Stokes equation is discretized directly even in the boundary cells in order to ensure the momentum conservation. Furthermore, the Navier-Stokes equation and the pressure Poisson equation in the boundary cells are constructed with the properly interpolated flux and pressure gradient. This treatment guarantees the consistency between the velocity and pressure fields in the boundary cells. The validity of the present method is assessed in some fundamental flows. It is found that the present method significantly improves the accuracy orders for the wall shear stress as well as the velocity compared to the voxel method and the conventional direct forcing immersed boundary methods. The results by the present method are also found to be less dependent on the Courant number due to the consideration of the consistency between the velocity and pressure fields in the vicinity of the boundary. The method is also applied to a large eddy simulation of a flow past a circular cylinder at a Reynolds number 3900. The flow fields predicted by the present method are found to be in good agreement with those of the experimental and numerical studies reported in the literatures.
An improved system of thermo-anemometer was developed. Temperature compensator was designed based on more accurate equations. Assumptions in the temperature compensation equation were minimized to make minute adjustment in each part of the circuit. An improved corrector circuit reduced the error of the velocity signal to be less than ±1% up to about 120°C. A new digital delay device was designed. Deviations from the specification of delay device could be adjusted by employing an automatic amplitude and mean voltage adjuster. Consequently, it can regenerate the original velocity waveform quite accurately, free from wave deformation and electrical noise up to 6kHz. Calibration time was reduced. The present device can obtain higher correlation terms in the thermal flow fields more accurately.
We have experimentally found the liquid film flow characteristics rising along the outer surface of a rotating cone, where the surface tension, Coriolis force, pressure gradient etc. work to maintain the liquid film on the surface of the cone. When the film flow goes up fully upward, the liquid film can not keep the filmwise condition and is eventually atomized into a mist flow. The mechanism can be used in a new device to atomize liquid and to generate the mist flow. In this research, we apply the new atomization mechanism to an oxygen mass transfer from the air to the water through the atomized water droplets. A dissolved oxygen concentration is measured for variations of rotation rates of the cone, and a volumetric mass transfer coefficient is calculated from the measured data. It is found that the mist flow is effective for the oxygen mass transfer judging from the obatained quite large oxygen transfer coefficient.
Homogeneous charge compression ignition(HCCI) combustion is promising a new combustion system reducing NOx and PM simultaneously without any penalty of fuel consumption. However, the operational range of the HCCI combustion system is limited because of some issues such as poor control of ignition timing and the excessive rate of pressure rise. In this study, a new combustion system based on an HCCI combustion process will be proposed. The combustion system has a pre-chamber in the cylinder head. On the combustion system, at first, ignition takes place in the pre-chamber, and then the burned gas ejected into the main chamber ignite the mixture in the main chamber. In the combustion system, the combustion process in the main chamber is aimed to go on same manner of the HCCI combustion process. This paper presents the concept of the combustion system and test results of some experiments.
Heat transfer experiment and pressure loss measurement have been done for the channel flow with aluminum fiber layers in order to investigate their effectiveness as an insertion device to improve the heat exchanger's performance. Aluminum fiber layers focused in this paper are composed of aluminum fibers of 100 μm in diameter, once laminated as a non-woven fabric and then hardened by diffusion bonding as a bulk insertion body, and are expected to attain large heat transfer enhancement due to their high thermal conductivity and fine heat spread structures. Since heat conduction anisotropy exists in such fibrous layers, mainly two orientations, that is, fibers axes parallel and perpendicular to the heat transfer wall, were tested as parameter in addition to the other parameters, such as the porosity and the bonding method to the wall. It was found that the aluminum fiber layers with their axes mainly perpendicular to the wall showed quite high heat transfer performance, that is, heat transfer of twenty times as large as that of the non insertion case was obtained, although quite large pressure loss was required.
A sophisticated technique to measure thermal properties using inverse heat conduction analysis was proposed and applied to paraffin or packed beds such as copper and metal hydride particles. The technique was extended to a finite hollow cylindrical domain with new accuracy assurance concepts about criterions deciding thermocouple positions in a sample and time period to be solved the inverse solution. The features of the technique are as follows. 1) Arbitrary heating boundary conditions without constant temperature or heat flux can be used. 2) Thermal diffusivity and heat conductivity are quickly and simultaneously obtained after acquisition of transient sample temperature histories and cumulative heat input during sample heating. 3) Estimation errors in thermal properties are assured within 5 %. This technique was successfully applied to measurement of hydrogen absorbed metal hydride bed.
Steam injector is a passive jet pump which operates without power source or rotating machinery and it has high heat transfer performance due to the direct-contact condensation of supersonic steam flow onto subcooled water jet, and it has been considered to be applied to the passive safety system for the Next-generation nuclear power plants. The objective of the present study is to clarify operating mechanisms of the injector and to determine the operating ranges. In this study, temperature and velocity distribution in the mixing nozzle as well as flow directional pressure distribution were measured. In addition, flow structure in whole of the injector was observed with high-speed video camera. It was confirmed that there were unsteady interfacial behavior in mixing nozzle which advances heat transfer between steam flow and water jet with calculation of heat transfer coefficient. Discharged pressure at diffuser was also estimated with a one-dimensional model proposed previously. Furthermore, it was clarified that steam flow did not condense completely in mixing nozzle and it became two-phase flow in throat and diffuser, which seemed to induce shock wave. From those results, several discussions and suggestions to develop a physics model which predicts the injectors operating characteristics are described in this paper.
We have developed a non-contact high speed viscosity sensing technique, laser-induced capillary wave (LiCW) method using pulsed volume heating laser of near-infrared wave length. The main idea of the present work is based on the capillary wave induced by volume heating, which behaves more physically simplified than the one induced by surface heating in decay process and has nanometer-scale amplitude even as relatively-small temperature rise. We have derived the new theory for the wave amplitude z (x, z) captured the physics of volume heating by giving the boundary condition of heat conduction into the depth direction. First, we compared the theoretical damping behavior of capillary wave for toluene by volume heating and surface heating. According to the proposed theory, the capillary wave induced by volume heating is formed by only the effect of the thermal expansion with having the negligible effect on the temperature dependence of surface tension. In addition, maximum temperature rise and wave amplitude of water and toluene, absorption length of them are extremely different from each other, was compared between volume heating with surface heating. As a result, it was confirmed that nanometer-scale capillary wave can be induced with the temperature rise of less than mK order by volume heating, which indicates that near-infrared wave length is more applicable to the thremophysical measurement technique as a heating light source. Finally, to demonstrate the validity of the new theory, we have measured viscosities and surface tensions of Newtonian liquids, which showed good agreement within ± 5 % from the reference values.
This paper presents an experimental investigation of motion characteristics of microbubbles rising close to a vertical plane wall. The study focuses on how microbubble behaviors change with the bubble diameter and along the wall for a constant bubble flow rate. Tap water is used as the working fluid, and hydrogen microbubbles are generated by water electrolysis. A particle tracking velocimetry technique is used to precisely measure the microbubble velocity, with which we found that the mean rise velocity of microbubbles is much faster than theoretically estimated rise velocity of single microbubble in stationary water. The mean bubble rise velocity increases with the bubble diameter and decreases in the downstream direction, and both are related to the balance between accumulation and diffusion of spatial buoyancy distribution near the wall. In particular, in the case of small microbubbles, bubble clouds are generated close to the wall and show intermittent roll-up motion in the wall-perpendicular direction (void burst motion). We discussed on this motion with various waveforms of the bubble rise velocity and bubble-bubble distance.
This paper demonstrates an inverse design to explore the possible best binary diffractive microlens which functions as a wavelength selective concentrator. EA-FDTD method in which evolutionary algorithm (EA) was incorporated with finite difference time domain (FDTD) computation was used for optimization of the lens structure. The present method successfully generated the selective concentrator structures within practical computation time. This implies that the present method has a potential to design complicated but desirable photonic microstructures which the conventional methods hardly design efficiently. In addition, effects of EA parameters such as mutation rate and crossover rate on convergence speed were tested. Transformation of the lens structure was monitored generation by generation. It was found that there was a suitable set of EA parameters for each generation steps to accelerate convergence. If EA parameters can be controlled properly at each generation steps, convergence speed becomes faster. It was also found that the best structure was obtained efficiently if the degree of freedom for the part of the lens structure was controlled properly during the generations. Those results will contribute to further development of EA-FDTD method.
Regulations governing diesel engine emissions have become more stringent from the viewpoint of environmental considerations and cannot be fulfilled simply by improving combustion efficiency. For removing diesel particulate matter (PM) from diesel exhaust gas, utilization of diesel particulate filer (DPF) is effective in the present technology. However, treatment of the collected PM in the DPF, that is, DPF regeneration is a problem. An ideal method is low-temperature regeneration using no catalyst. To overcome the problem, a treatment method of PM using ozone at a relatively low temperature is proposed. An experimental evaluation of the DPF regeneration is performed by regulating the exhaust gas temperature. From the evaluation, the DPF regeneration is achieved at the 180, 250 and 290°C. Among them, the highest performance of the DPF regeneration is achieved at 250°C with the efficiency of 7.10 g(PM)/kWh. The plasma energy required for DPF regeneration is only 0.24% of the power generated by the engine.