In studying the mechanism of aerodynamic noise generated by the body of an automotive vehicle, it is necessary to consider the effects of not only an object in the air flow, but also the body surface located close to the object on the noise. In the present paper, a square cylinder and a flat plate were employed as a test model, and aerodynamic noise in the far field and pressure fluctuations on the flat plate were examined by changing the distance between the square cylinder and the flat plate. A commercial, numerical simulation code based on Lattice Boltzmann Method was used to obtain the data on the flow structure and aeroacoustic properties. In parallel with this numerical simulation, an experiment was conducted with regard to this model. It was found from the experiment that the aeolian sound is generated by Karman vortices when the clearance between the square cylinder and the flat plate is 60% height or more of the square cylinder. In addition, the flat plate plays a role of an acoustic reflection surface. The effect of the flat plate on the far field noise level appears in the frequency region with St =0.25 or more, and an increase in the noise level by the acoustic reflection is about 5dB. Therefore, it is necessary to consider the non-compact acoustic characteristics to estimate the far field noise level.

抄録全体を表示

In recent years, the engine noise level has been reduced due to powerplant electrification seen in EVs and HEVs for improving fuel economy. At the same time, the wind noise has become more noticeable to the degree that may take away driver comfort. At Honda, we have verified that the contribution of wind noise increases as the speed of EV increases. Among different wind noise types, the high frequency aeroacoustic noise (1k-5kHz) is generated by the flow around the side mirrors and front pillars, and it is transmitted into the cabin through the side window glasses. However, the mechanism of noise generation and transmission has not been fully understood. In this research, our focus is to clarify control object for the reduction of the high frequency aeroacoustic noise into the cabin. We use a simplified noise transmission model in which the noise from a forward facing step mocking car pillar being transmitted into a box set as car cabin. By applying CFD and CAA analysis validated by wind tunnel experiments, we identify the contribution of convectional pressure fluctuations and/or acoustic sounds. Our analysis results show that the prediction values of the box sound pressure which takes account of convectional pressure fluctuations and acoustic sounds have a good agreement with the experiments, and the contribution of acoustic sounds is about 93% of the sound pressure in the box. Finally, we clarify that in case of high frequency aeroacoustic noise, the acoustic sounds induced by the flow around the forward facing step are directly transmitted into the box.

抄録全体を表示

To clarify the mechanism of acoustic radiation in flows around a cascade of flat plates, fluid structures and acoustic fields were elucidated by direct simulations. The simulations were mainly performed for flows around 5 parallel plates and the separation-to-thickness ratio s/d was 6.0. The freestream velocity was changed from 30 m/sec to 60 m/sec, and the acoustic resonance occurs between plates at the freestream velocity of 44 m/sec. At that velocity, the Reynolds number based on the chord length and the freestream velocity was 8.7×10⁴ and that based on the plate thickness was 5.8×10³. Computational results were validated by the experimental results performed in the present research. The computational results showed that large-scale vortices composed of fine-scale vortices are shed in the wakes of the plates independently of the acoustic resonance. When the large-scale vortex is shed from the upper or lower face, an expansion wave is radiated around the downstream edge of the upper or lower face, respectively. The compression wave is radiated around the downstream edge of the opposite face. The simulation for the flow around a single plate was also performed, and the results confirmed the above-mentioned acoustic radiation mechanism. For the flows around a cascade of flat plates, the sheddings of the vortices from neighboring plates are synchronized when the acoustic resonance occurs. It was also clarified that the mode of the synchronization is an anti-phase mode and the standing waves generated between plates are reinforced.

抄録全体を表示

In the present paper the attention is focused on effect of helical strakes around a finned tube on Aeolian Tone. The helical strakes are mounted spirally around a twist serrated finned tube. We measured the turbulence intensity in the wake, spectrum of velocity fluctuation, the coherence of velocity fluctuations in the spanwise direction and SPL. The Aeolian tone radiated from the finned tube with helical strakes at high Reynolds number was smaller than the case of no helical strakes. The helical stakes were effective to reduce the Aeolian tone radiated from the finned tube. However, the effect of helical strakes of small diameter on vortex shedding depended on the Reynolds number. The Karman vortex was clearly formed even in the case of helical strakes of small diameter at low Reynolds number. The existence of helical strakes of small diameter around a finned tube caused the increase of the periodicity of vortex shedding from a finned tube. The coherent scale of Karman vortex in the spanwise direction was larger than that of a finned tube without helical strakes.

抄録全体を表示

In the present study, we consider the air entrainment into a suction pipe which is vertically inserted down into a suction sump across a mean free-water surface. This configuration is often referred to as the “vertical wet-pit pump”, and has many practical advantages in construction, maintenance and operation. In particular, we focus our concern upon the critical submergence depth S_c, which is one of the prime and conventional indicators for the air-entrainment occurrence. By a systematic approach, we experimentally investigate the influences of kinetic and geometric parameters upon S_c. As the kinetic parameters, we consider the Reynolds number Re and the Weber number We, in addition to the Froude number Fr, on such a basis as Fr is not much larger than unity in many actual cases. As the geometric parameters, we consider back clearance X, sump breadth B and bottom clearance Z. All these parameters are non-dimensionalised by the outside diameter D and the intake velocity V_i of the suction pipe. As a result, we reveal the effects of such six parameters upon S_c. The We effect, namely, the surface-tension effect can be ignored at We > 12. And, the Re effect, namely, the viscous effect becomes negligibly small at Re > 3×10⁴. Under such conditions for We and Re, we could consider only the Fr effect, namely, the gravitational effect. Concerning the X/D and B/D effects, S_c/D attains the maximum at a certain X/D or B/D. On the other hand, the Z/D effect is monotonical, and becomes small at Z/D > 2.5. Some aspects of these geometric effects can be evaluated by a local-Froude-number effect on the basis of the global relation between S_c/D and Fr. And, the other aspects is necessarily considered to be related with the flow structure in the suction sump.

抄録全体を表示

We have investigated the behaviors of rod-like hematite particles in a simple shear flow as well as in an applied magnetic field, in order to clarify the dependence of the negative magneto-rheological effect on particle aggregation and orientational distribution of particles. The present Brownian dynamics method has a significant advantage that it takes into account the spin rotational Brownian motion around the particle axis in addition to the ordinary translational and rotational Brownian motion. The net viscosity is decomposed into three components and discussed at a deeper level and in more detail: these three viscosity components arise from (a) the torque due to the magnetic particle-field interaction and (b) the torque and (c) the force due to the interaction between particles. In contrast to the ferromagnetic rod-like particle that exhibits a single-peak-type orientational distribution, the present hematite particle has a linear-peak-type distribution with a much more gentle profile. Hence, the orientational distribution does not change significantly. However, this slightly reformed distribution makes a significant influence on the negative magneto-rheological effect. In a dilute suspension, the effect of the magnetic field strength on this negative magneto-rheological effect is qualitatively and quantitatively in good agreement with that obtained by the previous theory and Brownian dynamics simulations without translational Brownian motion. In a relatively dense suspension, the viscosity components arising from an applied magnetic field and the interaction between particles come to change rapidly since certain strength of magnetic particle-particle interaction, which is due to the formation of raft-like clusters.

抄録全体を表示

Spatial development of a turbulent field generated by a multiscale, fractal-grid is investigated in a wind tunnel. A square type (i.e., fractal elements with square shapes) fractal grid is placed at the inlet of the test section. The Reynolds number based on the length of the biggest grid bar is set to 5900. The spatial developments of cross-sectional profiles of turbulent kinetic energy K, dissipation rate of K, ε, and the advection, production, triple-correlation transport, pressure transport and viscous diffusion terms in the transport equation of K are measured by using a hot wire anemometer. The results show that in the upstream region, turbulence produced by the biggest grid bar is transported to the central and outward regions by the triple-correlation term T. In the decay region, there is low turbulence production downstream of the interior of the biggest grid bar, which is mainly transported outward rather than to the central region by T. The advection term is large and positive in the decay region whereas the pressure transport and the viscous diffusion terms are very small in the decay region.

抄録全体を表示

Aerodynamic noise with discrete frequencies is generated in an unexpected way from a slit between a flat plate, which is placed in parallel to the main flow, and a wedge that is placed normal to the flow. To study this phenomenon, we examined the discrete frequency noise generated by a subsonic air jet flow from a slit. It was found that the slit height and the plate length have considerable effects on the noise generation. In addition, the discrete frequency noise was generated only when the values of these parameters are within a limited range. Furthermore, it was found that the noise frequency is proportional to the jet velocity and is inversely proportional to the plate length. Finally, it was made clear from the present study that the interaction of the vortex, which is generated from the shear layer due to a wedge, with the wake of a flat plate is a major cause for the discrete noise due to a slit.

抄録全体を表示

Void fraction, gas-liquid phase lengths, and frictional pressure drop were measured for gas-liquid two-phase slug flow in circular horizontal microtubes. As the test tubes, fused silica capillary tubes with the inner diameter of 75, 100 and 200μm were used. In this experiment, tap water and compressed air were working fluids, and three kinds of mixer with different inner diameter and shape were used. The flow pattern of slug flow was observed by the high-speed camera connected to the microscope. Void fraction and pressure at the observation point were calculated using a correlation for liquid film thickness. It was found that the distribution parameter, C₀, in the drift flux model was related with capillary number based on the total volumetric flux. The void fraction was estimated well using the correlation of C₀. The dimensionless gas phase length was changed by the cross sectional area ratio of microtubes to mixers. It was observed that the frictional pressure drop, (dP/dz), was changed as the test section's diameter. In addition, (dP/dz) was correlated well with the Lockhart-Martinelli method and the unit cell model.

抄録全体を表示

In order to utilize the results obtained by numerical flow analysis in an effective way, a method to evaluate the pressure loss due to energy dissipation has been developed in this study. Specifically four types of two-dimensional sirocco fan with different values of blade thickness were simulated, and the evaluation method proposed here was applied to their results. As a result, the location, at which the pressure loss occurs, as well as the amount of pressure loss was made clear. In addition, by dividing the total pressure loss into four kinds of loss: i.e., the collision loss, the separation loss, the wall loss, and the wake loss, the cause of each loss was identified. Consequently, in the original sirocco fan the wake loss accounted for approximately 40% of the whole pressure loss between blades, while in the modified fan it reduced to 50% of the original fan. Thus this evaluation method using energy dissipation was found to be very useful to see the pressure loss, and can be extended to 3D without any difficulty.

抄録全体を表示

An effect of nozzle-bucket axial gap length on the unsteady force acting on steam turbine buckets was studied by using a new method to predict the unsteady force which is induced by a potential field interaction and a wake interaction between nozzles and buckets. The new method can separate the unsteady force into two components due to the potential field and the wake interactions by calculating the inviscid and the viscous flow solution, and the effect of nozzle-bucket axial gap length is obtained. A simple relationship between the nozzle-bucket axial gap length and the unsteady force was found to give easy and reliable prediction of the unsteady force.

抄録全体を表示

Three-dimensional numerical predictions are presented for a laminar flow through a square channel with a sudden expansion of area ratio 4 : 1 in the range of Reynolds numbers Re = 20 to 1,900, using an Open Source CFD Toolbox (OpenFOAM). In particular, the effect of Reynolds number variation on the bifurcation of flow structure downstream of the sudden expansion and the evolution of unsteadiness was investigated. The results showed that the whole flow field behind the sudden expansion could be classified into three different regions as Reynolds number varies. The steady flow region was in the range of Re ≤ 900, the unstable steady flow or unsteady disordered motion flow appeared in the range of 900 < Re ≤ 1,300, and the unsteady disordered motion flow was stable in the range of 1,300 < Re ≤ 1,900. The observation area of disordered motion could be determined from RMS value of streamwise fluctuating velocity on the flow center line. As the Reynolds number increased, the observation area moved to upstream direction and the maximum RMS value increased. The high speed flow cluster near the centerline was diffused toward the wall by the disordered motion. So that, the vortex structure in the recirculation zone was disturbed by the high speed flow and the reattachment points were moved to upstream direction. The development of the separating shear layer depended on the initial position of the disturbance that added to the diagonal corner or the wall centerline in the upstream flow region of sudden expansion.

抄録全体を表示

Labyrinth seals have the potential to cause rotordynamic instability induced by the fluid force of seal flows. We conducted a computational fluid dynamics (CFD) study to investigate the rotordynamic characteristics of the shroud labyrinth seal of a high-pressure steam turbine. We constructed a CFD model consisting of a 1-stage stator/rotor cascade and a labyrinth seal over the rotor shroud to properly evaluate the seal force of an actual steam turbine and numerically investigated the effects of different seal gaps (0.2-0.8mm) and fin configurations (stator fin/rotor fin). The predicted stiffness coefficients increased inversely with decreasing seal gaps up to a gap of 0.4mm, but a seal gap smaller than 0.4 mm resulted in a slight increase in stiffness coefficients. Through investigation of flow fields, it was found that this could be attributed to the difference in the distribution of circumferential velocity component inside seals and the corresponding fluid force distribution. A comparison between different fin configurations showed that the stator fin indicated a higher stiffness coefficient, but the rotor fin became unstable at a higher whirl frequency due to its lower damping characteristic.

抄録全体を表示

In order to improve the record of the javelic throw in the junior Olympic games, it is important to elucidate the flight characteristics of the “turbo-jav” used in this throwing event. Unlike the spear used in the javelin throw, the turbo-jav has four tail fins. Focusing on the effect of the fins, we investigate the flight characteristics of the turbo-jav, by means of wind tunnel tests, throwing experiments and numerical simulations. In the wind tunnel test, we measured the drag force, lift force and pitching moment acting on intact turbo-javs as well as turbo-javs with their fins cut, in a low speed wind tunnel at the flow speed up to 25 m/s. The experimental results showed that the drag and lift coefficients for the intact turbo-javs are larger than the corresponding values for the turbo-javs without fins. As the angle of attack increases from 0, the moment coefficients for the intact turbo-javs decrease from 0, whereas the moment coefficients for the turbo-javs without fins increase. In accord with this property for the pitching moment, the throwing experiments showed that intact turbo-javs fly stably with oscillating angle of attack around 0. The flying distance, the orbit and the variation of angle of attack for the intact turbo-javs launched by a launcher agree closely with the numerical simulation performed based on the wind tunnel test. A comparison of throwing experiments by an athlete and by the launcher suggested that the turbo-javs flying without rolling could reach farther than turbo-javs with rolling.

抄録全体を表示

A aerodynamic noise problem in an automotive air-conditioning multi-blade blower was investigated numerically. The preceding measurements of the noise from a multi-blade blower demonstrated that the noise having a peak at 60 and 110 Hz was generated when it was operated at a low flow rate (φ=0.09), whereas the noise broadly ranged from 400 to 800 Hz was detected for a high flow rate (φ=0.22). In order to investigate the aerodynamic mechanism that generates these noises, flow in the multi-blade blower was numerically studied with large eddy simulation. The results showed a large pressure fluctuation at the tongue in the low flow rate. Observing the flow near the tongue, it was found that a re-circulated flow from the scroll terminal collided with specific point on the tongue surface and an excited oscillatory flow was issued periodically. Due to this flow behavior, it was postulated that the noise at 60 and 110 Hz for the low flow rate condition was induced by the flow impingement against the tongue related to the self-excited oscillation. In contrast, for the high flow rate, a prominent pressure fluctuation was found at the tongue and the blade surface of the shroud side. The flow visualization indicated that vortices shed from the trailing edge, interacted with a main stream and collided with the tongue. In addition, vortices in the shear layer which was formed behind the bell-mouth intermittently convected into a passage of the blades. These vertical interactions are considered to cause a large pressure fluctuation at the tongue and between the blades, thereby attributed to be the noise source in the high flow rate condition.

抄録全体を表示

Electrochemical partial oxidation (EPOx) of methane can convert exhaust heat into electricity as much as difference between change of Gibbs free energy and change of enthalpy. To quantify recuperated heat and converted electric power of EPOx, we simulated the performance of EPOx in the microtubular Solid Oxide Fuel Cell (SOFC) using Gadolinium Doped Ceria as electrolyte. The quasi-two dimensional and non-isothermal model applied to this SOFC simulation, which consisted of three solid layers and two gas layers, considering with mass, energy and chemical species conservation equations as well as detailed electrochemical reaction. The simulation computed temperature and current density distributions, and evaluated energy flow in SOFC. The simulation code was validated by consistency between the simulation result of power generation using H₂ as fuel and the result of previous experimental report. The results showed that EPOx could convert 40% of theoretically recuperated heat into the electric power at the operation condition maximizing total regenerated heat.

抄録全体を表示

Experimental studies are carried out for Ethanol and DME flames using counterflow flame configurations. A comparative study has been conducted to investigate the extinction, autoignition, and flame structure of both fuels. Extinction, autoignition and flame structure measurements on nonpremixed flames are carried out using a counter flow burner. Concentrations of stable species are measured by removing gas samples from the reaction zone using a quartz microprobe and analyzing them in a gas chromatograph. The temperature profile of the nonpremixed flame is measured using a thermocouple. Numerical calculations are also performed using detailed chemistry at conditions corresponding to those used in the experiments to validate chemical kinetics.

抄録全体を表示

Experimental studies are carried out for oxygenated and hydrocarbon fuels flames using counterflow flame configurations. Oxygenated fuels include DME, ethanol, and butanol, and hydrocarbon fuels include propane, n-heptane, and iso-octane. A comparative study has been conducted to investigate the extinction, autoignition of all fuels. Studies on nonpremixed flames for extinction and autoignition measurements are carried out by injecting a fuel stream made up of fuel and nitrogen from one duct and an oxidizer stream made up of oxygen and nitrogen from the other duct. In the premixed flame configuration, the strain rate at extinction is measured over a wide range of equivalence ratio and oxygen concentration by injecting the premixed-reactant stream from one duct and nitrogen from the other duct. Numerical calculations are also performed for oxygenated fuels using detailed chemistry at conditions corresponding to those used in the experiments to validate chemical kinetics.

抄録全体を表示

Silicon nanoparticles (Si NPs) are expected to be high-value added materials for the next generation devices due to their size dependent optical and electronic properties. Previously, we have developed in-flight plasma CVD for synthesis of free-standing, size-tunable Si NPs with mean diameter of smaller than 10 nm using silicon tetrachloride (SiCl₄). As-produced, chlorine-terminated Si NPs are readily dispersed into organic solvent without surface modification, enabling a suspension of Si NPs to form silicon inks. By adapting appropriate solution process with silicon inks, Si NP containing devices can be fabricated without high vacuum dry process, which provides great flexibility for the development of Si NP based devices at low cost. This paper clarifies the operation regime of in-flight plasma CVD to produce desired Si NPs in terms of yield, size distribution, and crystallinity towards Si NP containing photovoltaics. Size distribution and crystallinity are controlled by residence time (or total flow rate) and input power respectively, while the yield is unambiguously determined by SEI, or specific energy input, which is calculated by the power (J/s) over the total flow rate (cm³/s). Moreover, gas analysis by mass spectrometer revealed that Si NP yield increased as SiCl₄ conversion, however, increase in HCl suppresses particle nucleation and amorphous silicon film deposition on the reactor wall became the main silicon loss pathway. Moreover, effect of H₂/SiCl₄ ratio was investigated for better understanding of Si NP formation mechanism.

抄録全体を表示

Energy harvesting is a process of extracting small amount of energy from ambient environment. Currently, electricity of μW range is enough to drive small electronic devices such as wireless sensors and networking devices. This study reports experimental result of a unique boiling energy harvester (BEH) which gains power from kinetic energy of boiling bubbles by means of piezoelectric devices. To verify working principle and characteristic, fundamental experiment has been conducted by a visible BEH with two types of piezoelectric materials (PVDF and PZT). As a result, BEH successfully generated electricity from boiling phenomenon and maximum power of 9.2μW was obtained when using PZT. Vibration characteristic was analyzed to determine optimal electric load to BEH. In addition, effect of boiling heat transfer area and cavity height on the performance was also examined toward further improvement. Finally, it was shown that BEH coupled with energy harvesting circuit demonstrated practical power generation.

抄録全体を表示

The effect of solid particles on the performance of oscillating flow heat pipe (O-HP) has been investigated using a visualized loop O-HP experimental system. The O-HP system was made of Al plate with 2×2 mm chase and flow inside the chase was visualized through transparent plastic cover. Al₂O₃ nano particles (30 nm nominal diameter) were dispersed into ethanol and water as working fluids of O-HP. Heat transfer performance of the O-HP was tested under concentration of 0 - 3.0 wt% of Al₂O₃ nanofluids at volume filling ratio of 50%. Experimental results show that the water-based nanofluids enhances the heat transfer performance of the O-HP; however, after repeating the operation, the heat transfer performance decreases as the nano particles deposits. Once the deposits occur, it is difficult to recover the performance without an external manipulation. The reason of the particle deposition was experimentally investigated. The results imply that the particle deposition can be caused by the rapid increase of local particle concentration during evaporation process in narrow chase. Furthermore, physical properties (viscosity, thermal conductivity) of the nanofluids were measured and the mechanism of existence of optimal particle concentration was discussed.

抄録全体を表示

Spatial distributions of local heat transfer coefficients on the bottom wall downstream of a backward-facing step in a duct flow have been measured by making use of a thermo-sensitive liquid crystal sheet. Heat transfer experiments and three components of the time averaged velocity measurements by 2D PIV have been done for wide variety of the flow Reynolds number ranging from laminar (1000) to turbulent (10000) flow regimes for the stepped duct having an aspect ratio of 16 and expansion ratio of 2. Obtained local Nusselt number changes not only in the streamwise direction but also spanwise direction, that is, the maximum Nusselt number always appears near the side wall region even if the flow Reynolds number is varied. Strong downward flow from the step induced along the side wall produces the secondary flow toward the central part of the duct and the area where this secondary flow exists agrees well to the area of the high heat transfer coefficient. The streamwise location of the maximum Nusselt number quickly moves closer to the step as the Reynolds number increases from 1000 to 2300, then it gradually moves to the downstream from 2300 to 10000. This streamwise location corresponds quite well to that of the flow reattachment for each Reynolds number. A pair of the vertical vortices having axes in the wall normal direction is found to be generated at the corner just behind the step near the side wall and this causes the minimum heat transfer there.

抄録全体を表示

The effective thermal conductivity of a metal hydride packed bed was calculated in consideration of the influence of expansion and contraction. The area-contact model was modified so that the porosity and the deformed factor change by expansion and contraction. The porosity was calculated using the experimental formula developed by direct observation. The deformed factor was calculated using assumption of a simple geometrical deformation caused by difference between a particle expansion and a packed bed expansion. The calculation results are consistent with the reported experimental value. The calculation results indicate that the effective thermal conductivity is greatly influenced by the porosity and expansion ratio rather than other parameters. Therefore, this calculation method incorporating the influence of change of packed bed states, which are the porosity and the expansion ratio etc., will be helpful to investigate of heat transfer enhancement on a metal hydride reactor.

抄録全体を表示

This study was made to establish a prediction method of turbulent burning velocity for stoichiometric hydrogen-oxygen-argon mixtures, with paying special attention to the importance of local burning velocity to turbulent burning velocity. In this study, stoichiometric mixtures with different dilutions, whose laminar burning velocity S_L0 was varied from 25 to 90 cm/s, were used under almost the same weak turbulence condition, so that the ratio of turbulence intensity u' to S_L0 becomes 0.4 to 2.0. Nitrogen was also used as a conventional dilution gas for comparison with argon. A two-dimensional sequential laser tomography technique was applied to obtain the relationship between the flame shape and the flame displacement in a constant-volume vessel. The local flame displacement velocity S_F of turbulent flames was quantitatively measured as the key parameter of local burning velocity. It was found that the trends of the mean values of S_F, S_F,m, against u'/S_L0 corresponded well to its turbulent burning velocities, but S_F,m did not increase infinitely with increasing u'/S_L0. The turbulent burning velocity also attempted to be predicted by some empirical equations. As a result, the predicted turbulent burning velocities, where the obtained S_F,m was employed instead of S_L0 as the local burning velocity, were in good agreement with experimental ones.

抄録全体を表示

Structures of stretched cylindrical diffusion flame with large curvature were investigated experimentally. Temperature distributions of fuel-diluent mixture/air flames were measured. Fuel was diluted with diluent gases (N₂, Ar and He) in order to set the Lewis number Le～1. Fuel (Air) was supplied from inside (outside) of the cylindrical flame. Thermal dissipation rates were calculated from measured temperature distribution. In the case of Le=1, the scalar dissipation rate is proportional to the thermal dissipation rate. Therefore, the stoichiometric scalar dissipation rate was evaluated using the maximum value of the thermal dissipation rates obtained from each temperature distribution. Generally, it is known that the scalar dissipation rate of the counterflow flat diffusion flame increases as the flame stretch rate increases. However, in the case of the cylindrical diffusion flames, the scalar dissipation rate has a maximum value or decreases with the stretch rate.

抄録全体を表示

The thermal boundary conductance between the highly ordered pyrolytic graphite (HOPG) and aluminum, nickel and cobalt metal thin films has been measured at room temperature. The work aims to gain fundamental understanding into the thermal boundary conductance through the metal/graphite layer interface, which is essential for thermal management of graphene-based electric devices. The thermal conductance was measured using the time domain thermoreflectance (TDTR), a technique widely used to measure the thermal transport at the nanoscale. The measurements identify the dependence of the thermal boundary conductance on the metal kind. Furthermore, the trend was found to agree with that of the metal-graphene binding energy calculated by the density functional theory. The results suggest that the adhesion strength plays an important role in determining the thermal boundary conductance between metals and c-axis oriented HOPG.

抄録全体を表示

The mechanism of fuel-NO formation and recycled-NO reduction in fuel-rich O₂/CO₂ combustion is studied by a flat CH₄ flame experiment and detailed kinetics modeling. When fuel-N and recycled-NO are introduced into a flat flame simultaneously, NO is significantly decreased due to an interaction between fuel-N and recycled-NO. Detailed kinetics modeling reveals this interaction mechanism in NO reduction. The difference of N₂ production from fuel-N and recycled-NO between air and O₂/CO₂ combustion is caused by CO₂ reactivity expressed by CO₂ + H → CO + OH (R1b). From the viewpoint of CO₂ reactivity, O₂/CO₂ combustion is suitable for low-NO_x combustion under fuel-rich condition where NH₃ remains in air combustion.

抄録全体を表示

In general, a turbulent flame is observed in most practical combustors. As for the premixed combustion, by controlling the equivalence ratio of fuel and air, it is possible to suppress the emission of harmful substances such as NOx. However, the turbulent combustion process is very complex. Since there is unsteady flame and flow field interaction, it is very difficult to understand the combustion phenomena. To handle such a complex situation, the numerical cost may be too large. Then, by experiments, it is important to investigate turbulent combustion phenomena by laser diagnostics. In this study, we conducted a PIV/OH-PLIF simultaneous measurement of turbulent premixed flames. To focus on the effect of swirl flow on flame structure and stability, we used a cyclone-jet combustor. The extinction limit as well as CO and NOx emissions was experimentally obtained. By changing the flow condition of cyclone nozzles, we discussed the flow field with turbulence and local flame extinction.

抄録全体を表示

This paper describes characteristics of moisture losses in low pressure steam turbine stages. Steam turbines operating under wet-steam conditions experience thermodynamic and mechanical losses caused by the presence of moisture. In order to predict the moisture losses, physics-based moisture loss models are developed. The moisture loss models are composed of six categories: supersaturation loss, condensation loss, acceleration loss, braking loss, capturing loss and pumping loss. These losses are calculated according to the wetness levels in the stationary blade & moving blade rows, and defined by the functions of steam properties and design factors of turbine stages. The loss models are compared to full-size low pressure steam turbine data for prediction of the overall moisture losses. The developed loss models are found to be effective for designing wet-steam turbines from the standpoint of calculation accuracy and calculation time.

抄録全体を表示

In the previous report, as a new application of Stirling cycle, it is proposed to apply the Stirling cycle to heat speaker which amplifies sound vibration by means of thermal energy. In present report, the characteristic of this speaker was discussed and applying this theory an experimental device of heat speaker was developed. Using this device, the frequency characteristic of heat speaker was investigated. As a result, the amplification factor by heating decreased with frequency, and at the level of above 85[Hz] the amplification factors were minus value. The result of decreasing amplification factor is likely caused by decreasing of the movement of displacer attached to the voice coil.

抄録全体を表示