-
Akira SHIMA
1990 Volume 56 Issue 522 Pages
259-265
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
-
Fumio YOSHINO, Ryoji WAKA, Toyohiko SUZUKI, Koji TITIKAWA
1990 Volume 56 Issue 522 Pages
266-273
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Direct strip casting is attracting attention as a new method of production of steel plates. This paper describes the results of an investigation on how a liquid such as water flows on a circular cylinder. Depending on the nozzle height and water velocity at the nozzle exit, a roll wave, stationary-centrifugal-force wave, long wavelength wave, ripple and dimples appear, and a range of mirror-like surface flow between the exit Reynolds numbers of the roll wave and a stationary-centrifugal-force wave exists. The variations of water depth in the circumferencial direction are also measured.
View full abstract
-
Hideo OHASHI, Yoichiro MATSUMOTO, Yasumasa ICHIKAWA, Takashi TSUKIYAMA ...
1990 Volume 56 Issue 522 Pages
274-279
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
An air-water two phase flow recirculating water test tunnel was constructed. This tunnel was designed and constructed to measure the performance of either an isolated hydrofoil or a cascade of hydrofoils under variable two phase condition. Void fraction measuring system using Point Electric Resistivity Probe was also developed. In this paper, the specification and the basic characteristics of the tunnel, such as the distribution of velocity, void fraction and bubble radius at the test section, are reported. As an example of test results, the distribution of void fraction behind an airfoil is shown.
View full abstract
-
Atushi OKAJIMA, Taroh NAGAHISA, Akira ROKUGOU
1990 Volume 56 Issue 522 Pages
280-288
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
By a finite difference method, flows around cylinders with rectangular cross sections of various width-to-height ratios of 0.4 to 8 have been computed in the range of Reynolds numbers of l00 to 1200. We have successfully simulated some interesting phenomena whereby flow around the cylinder of the B/H ratio of 2 changes to a fully separated flow at a Reynolds number of 800, while for the cylinder of the B/H ratio of 2.8, a fully separated flow changes to an alternate reattachment one at a Reynolds number of 1200 with the lapse of time, two components with different Strouhal frequencies appear in a flow field around the cylinder of the B/H ratio of 6 at a Reynolds number of 800, and the flow pattern critically changes when the B/H ratio is 2.8 and 6 at Reynolds numbers of 500 to 1200. It is clarified by a numerical simulation that the component with a high Strouhal frequency is induced by vortices separated from the trailing edges and that the low Strouhal component is due to the oscillation of flow over the side walls accompanied by a movement of separation bubbles. Finally, the computed results of flow patterns, base pressure, drag force and Strouhal number show a good agreement with the experiments.
View full abstract
-
Tsutomu HAYASHI, Fumio YOSHINO
1990 Volume 56 Issue 522 Pages
289-294
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
A new technical method is presented for quantitative evaluation of the drag and lift coefficient on a circular cylinder in a uniform shear flow. The displacement effect originating from a shift of the stagnation streamline, the asymmetry of the pressure distribution on the circular cylinder, and the increase in back pressure which is caused by differences in the turbulence intensity in the free stream and in the aspect and the blockage ratio are considered as analytical elements. It is concluded that by using this technique, the lift acts on the cylinder from the faster side to the slower side of the free stream.
View full abstract
-
Jun ITO, Noriyoshi SATO, Masaru TAKAHASHI
1990 Volume 56 Issue 522 Pages
295-300
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
The guide vanes and the impellers of turbomachinery, because they operate between the casing and the boss, are in shear flow due to the wall boundary layer. The reason why so many problems involving this kind of branch are left unsolved seems to be the lack of applicability of the existing methods caused by mathematical or numerical difficulties. This paper aims to propose a concise and precise method of solution for the three-dimensional wing in shear flow between two parallel plane walls, which is a mode function method for the derived simultaneous integral equation.
View full abstract
-
Hiroya YAMADA, Ikuo NAKAMURA, Shintaro YAMASHITA, Haruhisa YANO
1990 Volume 56 Issue 522 Pages
301-306
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Experimental results of the secondary flow and the turbulence properties of a three-dimensional jet injected from a hole located on the downstream side of a circular cylinder into a uniform flow with a wake or into a quiescent fluid are presented. A secondary flow, such that the cross-sections of the jet are elongated by it to the z-direction, is observed, and deformation of the jet section injected into the uniform flow are considerably larger than those of the jet injected into a quiescent fluid. The fluctuating velocity profiles of the jet injected into a quiescent fluid become a similar form at x/d≳ 63. In the jet injected into the wake, the turbulence intensity of the jet becomes smaller than those of the wake. It is clarified from the power spectra that the Karman vortex disappears as a result of the interference with the jet.
View full abstract
-
Ikuo NAKAMURA, Yorinobu TOYA, Shintaro YAMASHITA, Yoshinori UEKI
1990 Volume 56 Issue 522 Pages
307-313
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
This experimental work is concerned with the instability of several Taylor vortex flows between concentric cylinders in a symmetric end condition. The flow has several characteristic patterns : primary mode, normal secondary mode, and anomalous mode which has outward flow on either the upper or lower end plate and/or on both end plates of cylinders. It is found that the anomalous mode changes other anomalous modes or normal secondary modes or primary modes when the Reynolds number is gradually decreased. Consequently, bifurcation set of complex flows is presented with the bifurcation diagram of flows in a symmetric system.
View full abstract
-
Yorihide SEGAWA
1990 Volume 56 Issue 522 Pages
314-321
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Characteristics of wave motion on an elastic wall adjoined by one or two vortex layers which appear freqently in a flow of high Reynolds number were solved by a numerical calculation. The following matters are clarified. (1) A coupled system comprising the vortex layer and the elastic wall becomes stable as the distance between the vortex layer and the elastic wall decreases in a case where virtual mass ratio is large. (2) Stable waves on the elastic wall become unstable for a jet-type velocity distribution, and unstable waves on the elastic wall become stable for a wake-type velocity distribution, as the distance decreases in the above case.
View full abstract
-
Yorihide SEGAWA
1990 Volume 56 Issue 522 Pages
322-329
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Stabilities of the vortex layer and elastic wall adjoined to this layer are analyzed taking into account the presence of the Coriolis force of the system. The following are clarified. (1) As static characteristics, the system is concerned with the generation of buckling and a stationary vortex, and the vortex layer of IS type (velocity U
2 (near the wall) < velocity U
1(far from the wall)) is stabilized while the vortex layer of the NS type (U
2 > U
1) is destabilized according to the decrease in the distance between the vortex layer and the wall. (2) As dynamic characteristics, the system is concerned with the generation of flutter and free vortex (vortex shedding), and the coupled system of the vortex layer and the elastic wall is stable, as the distance decreases where the virtual mass ratio related to the wall is large.
View full abstract
-
Hiromu SUGIYAMA, Takakage ARAI, Hukuyasu ABE, Tetsuo TAKAHASHI, Kazuyo ...
1990 Volume 56 Issue 522 Pages
330-335
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
This paper is concerned with the internal structure of λ-type pseudoshock waves in a straight-square duct. The experiments were carried out for M
1=1.77, where M
1 Was the flow Mach number just ahead of the pseudoshock wave, using a two-component, dual-beam laser Doppler velocimeter (LDV) and double exposure laser holographic interferograms. The velocity distributions in the pseudoshock wave were measured in detail. The variations of velocity and boundary layers displacement thickness were clarified. Especially, the displacement thickness increased at the shock wave locations and decreased in the regions between the shock waves. The velocity distributions calculated by the density distributions, which were measured by using the holographic technique, were compared with the results from LDV. It was found that the central flow in the pseudoshock wave was closely related to the isentropic flow, while the flow near the wall was closely related to the adiabatic flow.
View full abstract
-
Toshio MIYAUCHI, Koichiro KAWANO
1990 Volume 56 Issue 522 Pages
336-341
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
The aim of this study is a clarification of the spanwise and streamwise coherent vortical structures with initial random fluctuations in the temporally growing turbulent shear flows. To achieve this purpose, large eddy simulation of a turbulent shear flow was conducted. As an initial condition, random velocity fluctuations were superposed on the mean velocity profile, and 17 cases having different random fluctuations were calculated. From the calculated results, the following conclusions were obtained. (1) In spite of the existence of initial random fluctuations, spanwise and streamwise coherent structures can be observed. (2) The temporal development of the momentum thickness with initial random fluctuations does not show the lump which is observed in the case of periodic excitation.
View full abstract
-
Hideharu MAKITA, Akiyoshi IIDA
1990 Volume 56 Issue 522 Pages
342-350
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Three-dimensional energy and dissipation spectra were experimentally obtained in a large-scale turbulence field. The characteristic scales of turbulent eddies, the energy-containing eddy, l
o, the largest locally isotropic eddy, l
G, and the eddy of maximum dissipation, l
D, were determined in the spectral profile. The energy spectrum became self-similar for k>1/l
G but the l
o scale was reduced in accordance with the streamwise decay of the turbulence energy. The l
D scale was proven to be more appropriate as the scale of the energy-dissipative eddy than the Taylor microscale, λ. The energy spectrum started to depart from Kolmogorov's -5/3 low at η/l
D∼0.15 and was most accurately approximated by the theoretical profiles with an exponential tail in the far dissipation range.
View full abstract
-
Hideharu MAKITA, Akiyoshi IIDA
1990 Volume 56 Issue 522 Pages
351-359
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
A simple model for turbulence eddy hierarchy was proposed to describe the Kolomogorov's local similarity theory in connection with the characteristic time and eddy scales of turbulence. In weakly anisotropic turbulence, the energy fraction of locally isotropic eddies to the whole turbulence energy increased downstream and exceeded 60% of the total energy at X/M=50 for Re
λ=460. Most of the energy dissipation occurred in the quasi-equilibrium wave number range. The strong turbulence became quasi-self-similar and the turbulence Reynolds numbers were kept constant for X/M >65, where the time scale of the decay of turbulence energy became greater than 4∼8 times as large as the characteristic time scale of the energy-containing eddy at the initial stage of the decay. The strong turbulence was confirmed to be capable of offering an ideal test field for practical large-scale turbulence phenomena.
View full abstract
-
Masafumi HIROTA, Hideomi FUJITA, Hajime YOKOSAWA
1990 Volume 56 Issue 522 Pages
360-366
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Production and convection of turbulence kinetic energy have been examined in detail for turbulent flow through square ducts with rough walls. Three kinds of ducts- a smooth-walled duct, a duct with one rough wall, and one with two facing rough walls-have been used for the experiment. Repeating-rib-type roughness elements have been used to roughen the walls. It was found that the turbulent shear stresses play the most important part in the production of the turbulence kinetic energy. Contribution of the turbulent normal stresses also has great influences on the energy production near the rough walls, although the sign of their contribution changes depending on the relative location to the roughness elements. The turbulence energy convection by secondary currents is considerably smaller than the energy production and does not affect the balance of turbulence kinetic energy in the ducts.
View full abstract
-
Hideharu MAKITA, Takao IWASAKI
1990 Volume 56 Issue 522 Pages
367-373
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
The Reynolds stress model was investigated by analyzing the decay process of a homogeneous turbulence field with extreme axisymmetric anisotropy. The aspect of the change in the model constant of the destruction term in the transport equation of the turbulence energy dissipation rate was assured to vary with the degree of anisotropy defined by the invariant II of anisotropy tendor. The time scale of the decay of turbulence energy was not proportional to the time scale of the energy-containing eddy, L
1/u
1. Rotta's constant for the pressure-strain term was assumed fo change discontinuously as the invariant II decreased in the process of return to isotropy. Better universality was afforded to the model of the pressure-strain term when the anisotropy was assumed to be relaxed with the time scale L
1/u
1.
View full abstract
-
Masanori KANO
1990 Volume 56 Issue 522 Pages
374-381
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Two kinds of turbulence models, a k-ε model of Launder et al. using an anisotropic expression for Reynolds stress tensor, and a simple improved form of a classical Reynolds-stress model, have been applied to turbulent channel flow. The performance of both models has been investigated. A feature of the Reynolds-stress model is in the form in which the pressure-strain correlation terms give values on the wall compatible well with experimental data. The computed results are compared with experimental data. These results show that the calculated mean velocity profiles and turbulence intensities by the k-ε model are in good agreement with experimental data, while the calculated turbulence intensity in the x
1 direction by the Reynolds-stress model is predicted insufficiently. Therefore the pressure-strain correlation terms in the Reynolds-stress model are investigated.
View full abstract
-
Koich NAKABAYASHI, Osami KITOH, Hiroshi IWATA
1990 Volume 56 Issue 522 Pages
382-387
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
In this report, the discrepancy on the Law of the wall described in the previous report has been clarified. First of all, shear stress distribution has been theoretically and experimentally investigated. The relationship between velocity and shear stress gradients is considered in order to learn about the validity of the similarity laws on the velocity distributions, using the mixing length model. Karman constant x and damping-length constant A
+ proposed by Van Driest are obtained for a parameter a
sv/u
*3, and discussed in comparison with the moving equilibrium condition. Townsend's linear stress model is also discussed in the approximate gradient layer.
View full abstract
-
Hideharu MAKITA, Koji SASSA, Akiyoshi IIDA, Masaaki SHIGEYAMA
1990 Volume 56 Issue 522 Pages
388-395
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Particle diffusion from a point source was experimentally investigated in a homogeneous and quasi-isotropic large-scale turbulence field. The particle concentration was measured by a laser counting system. The flow visualization showed the meandering of a particle plume and the diffusion coefficient reached about 130cm
2/sec in the strong turbulence. The plume dispersed downstream in proporti on to t in the strong turbulence and to t
1/2 in the weak turbulence. The results were well desctibed by the Taylor's theory. The empirical relation for the plume spread as a function of the travel time assured that the turbulence generator made it possible to simulate the atmospheric diffusion in a wind tunnel with conventional size.
View full abstract
-
Takeshi KARO, Hitoshi NAKATANI
1990 Volume 56 Issue 522 Pages
396-401
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
A formulation of unsteady forces acting on bodies in three-dimensional, incompressible and inviscid flow is performed with a discrete vortex model. The flow field is determined by the number of discrete vortex filaments (vortex segment) as are the three-dimensional bodies. The total impulse acting on bodies in a certain time interval is calculated by applying the momentum theorem to control the volume which encloses the bodies and all vortex segments and which moves with the fluid. The forces are obtained by taking the time average of the total impulse and calculated by using the state of positions and the strengths of vortex segments at the beginning and the end of the time interval. This formula is applied to the calculation of the unsteady forces acting on a delta wing in uniform and nonuniform flow. As a result, it is found that this formula is very easy and effective to calculate unsteady forces with the three-dimensional discrete vortex model.
View full abstract
-
Katsuhiko SAGAE, Makoto KOIZUMI, Masanori YAMAKAWA
1990 Volume 56 Issue 522 Pages
402-409
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
A numerical method using a finite element approach has been developed to analyze unsteady flow of power-law fluids with free surfaces. In this paper, we propose a technique for calculation of a moving free surface; the method uses marker particles and volume of fraction, which is the occupancy rate of fluid in an element. The calculation result obtained by the present method for an L-shaped cavity flow was comparable with that obtained by the mesh rezoning method. It was within 5% difference in the calculation of location of the moving free surface. In the other examples, there are shown to calculation results of a moving free surface by the injection molding of two box-shaped cavity flows. Lastly, a calculation result for three-dimensional flow in a concave lens is shown.
View full abstract
-
Kenichi ICHINOSE, Hiroshi TOKUNAGA, Nobuyuki SATOFUKA
1990 Volume 56 Issue 522 Pages
410-416
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
A new direct simulation scheme using the eighth order accurate difference method is presented for three-dimensional instability of plane Poiseuille flow. The vorticity transport equation is transformed into a set of ordinary differential equations in time by using modified differential quadra-ture (MDQ) method for spatial discretization. The time integration is carried out by Runge-Kutta-Gill (RKG) scheme. The Poisson equations with respect to the vector potential are also discretized by MDQ method and solved by the iteration method. The multiple grid method, in addition, is employed to accelerate its convergence. The present method is compared with other method based on the primitive variables for certification. And further, our results of the amplitude decay and oscillation frequency of the disturbance, agree with those of pseudo-spectral method by Orszag, and the accuracy and efficiency of the present method is shown.
View full abstract
-
Hideo KOGUCHI, Shoji TAKAMURA, Toshio YADA
1990 Volume 56 Issue 522 Pages
417-424
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
In this paper, the behavior of the meniscus between two circular parallel plates in negative squeeze motion is theoretically and experimentally investigated. According to the previous study on two tilted rectangular plates, a local capillary number for circular liquid film is newly introduced and the theoretical relation between the local capillary number and the nondimensional wave number of disturbance agrees well the experimental one. It is however found that the nondimensional wave number of the experiments deviates from the theory in the case of high viscous liquid. The growth process of disturbances into fingers is precisely investigated by measuring the tip curvature of the fingers. The simple relationship between the ratio of the curvature to the thickness of the liquid and that of the curvature to the position of the tip can be found experimentally.
View full abstract
-
Yasushi IDO, Takahiko TANAHASHI
1990 Volume 56 Issue 522 Pages
425-432
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Assuming that the homogeneous electric field, magnetic field and flow direction were orthogonal to one another, a constant pressure gradient flow between two parallel plates was analytically dlscussed. A set of the system equations for conducting magnetic fluids considering the internal rotation which had been proposed by the authors in a previous report was used. Physical quantities, such as velocity, angular velocity of fluid and suspended particles, were theoretically obtained. The constant pressure gradient flow of conducting magnetic fluids was characterized by some dimensionless parameters (magnetic effect parameter, micropolar effect parameter, electric effect parameter efc.). Effects of these dimensionless parameters on the conducting magnetic fluids flow were invesitigated in detail. Furthermore, the limit of the range for the wall surface coefficient was given when applled electric and magnetic fields existed.
View full abstract
-
Hiroo OKANAGA, Yasushi YAMAMOTO, Takahiko TANAHASHI
1990 Volume 56 Issue 522 Pages
433-440
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
The representative finite difference methods and explicit time-marching methods, which have been ploposed for high-Reynolds-numbers flow and for vectorization, are compared by solving the rotating cone problem of the advective equation. The convective term is approximated by the second order central, QUICK, the third-order upwind and the fourth-order central difference methods. The high-order time-marching methods are used, e.g. the second-order or the third-order Adams-Bashforth method, the Predictor-Corrector method and the four-stage Runge-Kutta method. The QUICKEST method and LEITH method for the high-order accuracy calculation to the discretization of space and time marching, respectively, are also discussed.
View full abstract
-
Yasuhiro MATSUDA
1990 Volume 56 Issue 522 Pages
441-447
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
The unstationary convection-diffusion equation is one of the most important equations in heat and fluid analysis. The new effective numerical nethod was proposed for this equation by an error analysis technique based on the Fourier series. First, the conventional numerical methods of Finite-element method (FEM) and Finite-difference method (FDM), were revised by an error analysis technique combined with an analytical solution. Then, all of these numerical methods were evaluated from the standpoints of phase error and dissipation error. Twelve conventional methods were investigated (3 by FEM and 9 by FDM) and five methods were newly developed (1 by FEM and 4 by FDM). There were eight implicit methods and nine explicit methods. The moditied Galerkin method was proven to be the most efficient numerical method among these 17 methods. We confirmed this result through several numerical experiments.
View full abstract
-
Itsuro HONDA, Hideki OHBA, Tetsuji NAKIYAMA, Yuji TANIGAWA, Yukitoshi ...
1990 Volume 56 Issue 522 Pages
448-453
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
A driven cavity flow of a two dimensional incompressible viscous flow is calculated by finite difference method. A generalized coordinate system is used so that a sufficient number of grid points are distributed near the wall by Thompson's method. In order to satisfy continuity of mass flow rate, Poisson equation of the pressure which is derived from the MAC method is solved. And the three kinds of the third order upwind scheme are applied for the convective term in the Navier-Stokes equation. Furthermore, all calculations which employed the vectorized implicit time marching scheme are carried out on a supercomputer. As the results of computation, differences of the three schemes in the general coordinate at high Reynolds number are shown in this paper.
View full abstract
-
Ryooji IMAI, Yoshinobu TSUJIMOTO, Okitsugu FURUYA
1990 Volume 56 Issue 522 Pages
454-459
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Cavitating flows through a centrifugal impeller were analysed by using a singularity method. The location of cavity surface was determined by an iterative method and the cavity surface boundary conditions were applied exactly on the surface. The closed cavity model of the present calculation gives a thicker face cavity and a thinner back cavity as compared with the results of linear cavity models which assume that the boundary condition can be applied approximately on the vane surface. In case of back cavities, which have smaller thickness, the impeller characteristics are close to those obtained by using the linear cavity model ; higher head was obtained for face cavities. Models assuming a cavity wake with constant displacement thickness or a wake filled with dead water gave significantly lower head compared with that obtained by the closed model. Thus, it was found that, for centrifugal impellers, the cavity wake has a large effect on the characteristics.
View full abstract
-
Makoto YAMAMOTO, Chuichi ARAKAWA, Tetsuo TAGORI
1990 Volume 56 Issue 522 Pages
460-466
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Lobe mixer is a special nozzle equiped in a turbofan engine to increase the thrust efftciency and suppress the noise. At present the design depends on systematic experiment because the flow field is too complex to be represented with an eddy viscosity model. Although it is neccesary to use a Reynolds stress model(RSM), conventional ones have not taken account of 3 dimensional application. We extended Launder-Reece-Redi model and constructed a new low Reynolds number type RSM. To determine unknown constants and to confirm the ability 2 dimensional channel flow was calculated. The agreement between the calculation and the experiment is satisfactorily good. Then we applied it to the lobe mixer and showed that it can reproduce the flow qualitatively.
View full abstract
-
Shigeru MURATA, Nobuyuki SATOFUKA, Tadashi KUSHIYAMA
1990 Volume 56 Issue 522 Pages
467-474
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
A numerical method for computing two-dimensional incompressible flows is presented. The steady vorticcity-velocity equations are discretized with the central finite difference approximation and the resulting equations are solved by using the Multi-Grid method with the Group-Explicit-Iterative method. The driven flow in a square cavity is used as a test problem. Calculations are carried out for the case with Reynolds number as high as 3200 on 257×257 grid points. The computed results are in good agreement with those based on the vorticity-stream function equations, and it is found that large residuals of the numerical solutions for the equation of continuity and the defining equation of vorticity occur near the corners on the moving surface.
View full abstract
-
Takaaki SHIZAWA, Shinji HONAMI, Tsuyoshi SAITO
1990 Volume 56 Issue 522 Pages
475-482
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
This study discusses the development of a new flying hot-wire system with a reciprocating sled. A hot-wire probe attached to the sled driven by an AC servomotor has a maximum bias velocity of 5 m/s in the spanwise z-direction across the stream (in y-z plane). This flying hot-wire system has the advantage that two velocity components U and W can be obtained by processing the output voltages from a reciprocating normal hot-wire. Measurements of the longitudinal vortices embedded in the turbulent boundary layer were performed to verify the flying hot-wire system. The present system is a useful device for studying the large-scale motions of the vortices.
View full abstract
-
Susumu SAITO, Toshimichi SAKAI, Shin-ichi ISHII, Kenjiro HARA
1990 Volume 56 Issue 522 Pages
483-488
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
A mechanism to accelerate hot water in a pipe by boiling of liquid freon injected in the hot water was developed with the final objective of driving a Pelton wheel. Convergent, convergent-parallel and two convergent-divergent nozzles were used as the two-phase nozzles. The effect of the nozzle shape and the location of the freon injection were investigated. The temperature characteristics of water and amount of freon injection were measured. The convergent-divergent nozzle showed good results in energy conversion. At high temperature, some results showed higher values than theoretically estimated maximum values which were derived by considering only the boiling of liquid freon. This phenomenon was assumed to be a result of vaporization of water by agitation by freon gas at the nozzle, and this was confirmed experimentally by injecting air instead of liquid freon.
View full abstract
-
Hisakuni TAKENAGA, Yasuo KURAMOTO, Seiji FUKUDA, Takuji FUJIKAWA
1990 Volume 56 Issue 522 Pages
489-494
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
A Series of tests, including nozzle single tests, blade single tests and single-stage turbine tests, are conducted to study the practical applications of hot water turbines. Hot water turbines are considered to be very useful for waste heat recovery plants and geothermal power plants. In this report, single-stage hot water turbine performance data are presented. A hot water turbine performance estimation method covering both nozzle and blade performance is introduced, and this calculation is shown to accord well with the turbine performance test results. Further, it is confirmed both theoretically and experimentally that a high nozzle inlet pressure and assist steam improve turbine performance.
View full abstract
-
Yukimaru SHIMIZU, Takaya YOSHIKAWA, Shinji MATSUMURA
1990 Volume 56 Issue 522 Pages
495-501
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Output of horizontal axis wind turbine is improved by tip vane. The maximum improved value of tip vane turbine is 1.25 times larger than non-tip vane turbine output. The paper describes the relationships between turbine performance and configuration of tip vanes. Many tip vanes were used to improved the performance. It is found that V and S type tip vanes are vesy useful to improve the performance. Also, dimensions of V and S type vanes were investigated in details.
View full abstract
-
Yukimaru SHIMIZU, Takaya YOSHIKAWA, Kimihiko KAJIMOTO
1990 Volume 56 Issue 522 Pages
502-507
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
This paper presents the experimental analysis of power augmentation effects of horizontal axis wind turbine using a tip vane. The flow visualization method is used to analyze the augmentation phenomenon. The flow around a tip vane, tip vortex and flow patterns on the blade surface are visualized. The following results are described. (1) Strength of tip vortex, induced drag and turbulence on the blade tip can be diminished by the tip vane. (2) A tip vane can generate the diffuser effect.
View full abstract
-
Mitsuhiro FUKUTA, Tadashi YANAGISAWA, Takashi SHIMIZU, Yoshihiro MORIT ...
1990 Volume 56 Issue 522 Pages
508-513
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Since a compressor used for an automotive air conditioner is driven by an engine, refrigerating capacity will change according to the speed of the engine. While wobble plate compressors achieve continuous capacity control by changing the slope of the wobble plate, rotary compressors can perform only a few stage capacity control by opening or closing bypass ports. We developed a rotary compressor with a continuous capacity control mechanism which is activated by rotating a bypass port in a front plate, and evaluated its performance. The experimental compressor showed a high coefficient of performance under operating conditions which simulated practical conditions of automotive air conditioners, though its performance slightly decreased due to the re-expansion loss with the increase of the rotating angle of the bypass port.
View full abstract
-
Tsutomu NOZAKI, Kenji NAKANISHI, Yasushi NAGASAKI, Takahiro FUKUTANI
1990 Volume 56 Issue 522 Pages
514-517
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
A proposed sand collector having a small injection at the collecting port can be applied not only for gases but also for liquids as a carrier and can transport many kinds of grains, sands and other relatively small particles. Several variables affect the performance of this device. In this paper, the effects of the ratio of the flow rate of injection to the flow rate of suction on the concentration of glass beads are discussed experimentally using the glass beads and liquid two-phase flow. As a result, it is clarified that there exists a most provable ratio of the flow rate giving the maximum concentration of glass beads
View full abstract
-
Yasuhiko SAKAI, Ikuo NAKAMURA, Hiroyuki TSUNODA, Takehiro KUSIDA
1990 Volume 56 Issue 522 Pages
518-522
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
For the purpose of investigating the mixing process and reacting diffusion of multiple species in turbulent flows, a system of measuring the simultaneous concentration fluctuation of each species is developed using the light absorption spectrometric technique. This system makes it possible to measure instantaneous concentration of three species, which have light absorption bands between visible and near-infrared rays (wavelength 300∼2000 nm). An optical glass fiber with the diameter of 0.5 mm is used to introduce the light into a sampling part whose volume is about 4×10
-4 cc. The quantitative measurements are documented in a turbu1ent jet of the mixture of two species with the premixed ratio of 1.0. The measured concentration fluctuation properties for each species show good agreements with the past data, so it is ascertained that the present system enables one to measure simultaneously the concentration of two species.
View full abstract
-
Kenzou WADA, Takuya SUMIYOSHI, Nobuo HAYANO, Toshihiro KATAYAMA
1990 Volume 56 Issue 522 Pages
523-528
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
In the present report, dust collection performances in a fixed bed dust collector and in a cross-current moving-bed dust collector are experimentally investigated. In the experiments, coke is used as a packed medium and fly ash is used as dust. Taking account of the effect of the specific deposit of dust on the collection efficiency, the collection efficiency of a unit layer is expressed by an empirical expression. Simulations of the collection efficiency in the fixed bed dust collector are performed using the empirical expression. In the moving bed, simulations are performed by assuming the shape of the air stream. The results show that the dust collection efficiency is satisfactorily estimated by applying the expression in the fixed-bed and by using the model of the air stream in the moving-bed.
View full abstract
-
Toshio AIHARA, Joo-Kyun KIM
1990 Volume 56 Issue 522 Pages
529-535
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
The physical properties of a fluid vary drastically near its critical point. In this paper, heat transfer due to an axisymmetric laminar impinging jet onto a flat wall of uniform temperature is studied numerically, taking into account the pressure-and temperature dependence of all physical properties of supercritical carbon dioxide. Numerical solutions are obtained for the jet Reynolds numbers 500-2000, the jet pressure 7.4-8.4 MPa, and the temperature difference between the jet mouth and flat wall from 10 K to 150 K. When the jet temperature is very close to the pseudocritical point at a jet pressure of 7.5 MPa and lower, the distribution curve of the local Nusselt number has two or three peaks in the vicinity of the stagnation point, R<1; and the maximum heat flux does not occur at the stagnation point.
View full abstract
-
Atsumasa YOSHIDA, Yasuhiro NAKA, Tsutomu OHKITA
1990 Volume 56 Issue 522 Pages
536-540
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
Thermophysical and kinetic properties of the LaNi
5-H
2 system are investigated for the practical utilization of hydriding alloys. The measurements of the thermophysical properties are carried out by a transient hot-wire method along the P-C isotherms. The effective thermal conductivities of the system increase not only with the increase of the hydrogen gas pressure but also in the plateau region with the increase of the composition of the metal hydride. The specific heats of the metal hydride increase with the increase of the composition. The kinetic properties are measured under the isobaric and the isothermal conditions. The derived intrinsic chemical reaction rates indicate the difference in reaction mechanism between the absorption and the desorption processes.
View full abstract
-
Hidetoshi MASUDA, Seizi SASAKI, Mitsuo HIGANO, Hiroshi SASAKI
1990 Volume 56 Issue 522 Pages
541-546
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
A new principle for performing a simultaneous measurement of the specific heat, c, and the total hemispherical emissivity, ε
h, of metals is proposed. The measuring technique used is not the resistive self-heating method by electric current, but essentially the transient calorimetric method which is usually used only for measuring the ε
h. On this principle, two kinds of specimens need to be prepared; one is a specimen made of the material to be measured, and another is a compound specimen made of the above-mentioned material and a standard one. From the experimental results for these two specimens, the c and the ε
h values can be determined simultaneously. Constantan and copper were selected, respectively, as the test material and the standard, and measurements were performed in the range of 380-720K. The method appears suitable for performing a simultaneous measurement of c and ε
h of metals, particularly for measuring ε
h of alloys of which the specific heat data are lacking.
View full abstract
-
Tadao HARAGUCHI, Ryohachi SHIMADA, Satoshi KUMAGAI, Toshiro TAKEYAMA
1990 Volume 56 Issue 522 Pages
547-552
Published: February 25, 1990
Released on J-STAGE: March 28, 2008
JOURNAL
FREE ACCESS
The surfaces of dropwise condensation coated with noble metals are more expensive than those coated with polymers. In this paper we proposed the use of polyvinylidene chloride coatings as the promoter and investigated the effect of the thickness of the coatings to promote dropwise condensation of steam. Four kinds of thicknesses, 0.05μm, 0.10μm, 0.25μm and 0.50μm were examined. The surface was heat-set for 83 hours after each coating. Then each surface was evaluated for its ability to promote and sustain dropwise condensation of steam. The 0.05μm coating showed dropwise condensation as high performance of heat transfer as gold or oleic acid promoter surface. All thicknesses of film showed no signs of deterioration of promotion, for example the coating did not peel or crack even after heat transfer tests of transition of dropwise-filmwise-glacial condensation. The 10μm thickness of film was visually evaluated in its ability to promote and sustain dropwise condensation of steam for a long term. While exposed to steam at atmospheric pressure, good appearence of dropwise condensation was observed over a priod of 8076 hours.
View full abstract