Transactions of the Japan Society of Mechanical Engineers Series B
Online ISSN : 1884-8346
Print ISSN : 0387-5016
Volume 65, Issue 635
Displaying 1-49 of 49 articles from this issue
  • Atsushi OKAJIMA
    1999 Volume 65 Issue 635 Pages 2190-2195
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Download PDF (728K)
  • Atsushi OKAJIMA, Sumiaki OHTSUYAMA, Toshio NAGAMORI, Tomohito NAKANO, ...
    1999 Volume 65 Issue 635 Pages 2196-2203
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Flow induced in-line oscillation of a structure with a circular or rectangular section was experimentally studied by free-oscillation tests in a wind tunnel. Detailed data of response amplitude of various cylinders have been acquired in a wide range of the reduced mass-damping parameter (Scruton number), by free-oscillation tests of two-dimensional cylinder spring-mounted to oscillate as a rigid cylinder. Flow induced in-line oscillation of the cylinder is certified to have two types of excitation mechanism; one is the so-called wake breathing effect accompanied with a symmetric vortex street and the other is the vortex excitation at a half of the critical reduced velocity. Also a fixed splitter plate was mounted behind a cylinder to eliminate the vortex shedding of alternate vortices from the cylinder, after Aguirre's experiments, and the results were compared with those without a splitter plate. As a result, we can classify the flow induced in line oscillation of the circular or rectangular cylinder. For a circular cylinder and a square one, two types of excitation phenomena appear in each reduced velocity region, the thin rectangular cylinder with the side ratio b/h=0.4 suffers from the vortex excitation, and the elongated rectangular cylinder with the side ratio b/h=2.5, is subjected to the wake breathing effect accompanied with a symmetric vortex street. It is found that fluidelastic characteristics of two types of in-line oscillations are quite different; i.e., the response amplitude are very sensitive to the reduced massdamping parameter, during the in line oscillation with a symmetric vortex street, while the oscillation of vortex excitation are rather insensitive.
    Download PDF (888K)
  • Hiromitsu HAMAKAWA, Tohru FUKANO, Masashi ARAGAKI, Eiichi NISHIDA
    1999 Volume 65 Issue 635 Pages 2204-2211
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    The periodic phenomena of velocity fluctuation in the near wake of a circular cylinder with fin were experimentally investigated. In the present paper the discussion is focused on the effects of the shape and the pitch of the fin around a circular pipe on the periodic phenomena. The distributions of the mean velocity, the intensity of velocity fluctuation and its spectrum were measured as well as the coherence of a shedding vortex in the spanwise direction. As a result it was made clear that Karman vortex shedding was clearly formed even in the case of a cylinder with complicated fin. The increase in the number of fin effectively causes the increase in the equivalent diameter. The coherent scale in the spanwise direction is larger in the case with fin compared with that of the simple circular cylinder with the same diameter. A correlation of the equivalent diameter and the Strouhal number are also proposed.
    Download PDF (869K)
  • Tsutomu KAWAMURA, Toshitsugu NAKAO, Masanori TAKAHASHI, Masaaki HAYASH ...
    1999 Volume 65 Issue 635 Pages 2212-2220
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Fluctuating lift and drag forces by Karman vortex shedding and pressure distributions around a stationary circular cylinder were measured in a water cross flow at supercritical Reynolds numbers. The surface roughness of the cylinder was a mirror-polished surface. Turbulence intensities of the incoming filow were changed from 1% to 13% in order to investigate the effect on vortex shedding characteristics. As the turbulence intensity increased in supercritical region, flow separation location of the boundary layer on the cylinder surface moved forward, Strouhal number decreased from about 0.48 to 0.29 and the fluctuating lift force by Karman vortex shedding increased.
    Download PDF (964K)
  • Tsutomu KAWAMURA, Toshitsugu NAKAO, Masanori TAKAHASHI, Masaaki HAYASH ...
    1999 Volume 65 Issue 635 Pages 2221-2229
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Vortex-induced vibrations were measured for a circular cylinder subjected to a water cross flow at supercritical Reynolds numbers for a wide range of reduced velocities. Turbulence intensities were changed from 1% to 13% in order to investigate the effect of the Strouhal number on the region of synchronization by symmetrical and Karman vortex shedding. The reduced damping of the test cylinder was about 0.1 in water. The surface roughness of the cylinder was a mirror-polished surface. Strouhal number decreased from about 0.48 to 0.29 with increasing turbulence intensity. Synchronized vibrations were observed even at supercritical Reynolds numbers where fluctuating fluid force was small. Reduced velocities at which drag and lift direction lock-in by Karman vortex shedding were initiated decreased with increasing Strouhal number. When Strouhal number was about 0.29, the self-excited vibration in drag direction by symmetrical vortex shedding began at which the frequency ratio of Karman vortex shedding frequency to the natural frequency of cylinder was 0.32.
    Download PDF (1013K)
  • Takaaki SAKAI, Koji IWATA, Akira YAMAGUCHI, Masaki MORISITA, Kenji OGU ...
    1999 Volume 65 Issue 635 Pages 2230-2236
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Experiments have been conducted on vortex-induced vibrations of a circular cylinder in high Reynolds number (1×105<Re<3×106) water flow. Experimental results of displacements in drag and lift directions by resonant vortex-induced vibrations were compared to evaluate the effects of Reynolds number, the damping ratio and the tapered shape on them. As the results, these effects were confirmed on the criteria of the incipient of resonant vortex-induced vibration and its displacement.
    Download PDF (775K)
  • Koji IWATA, Akira YAMAGUCHI, Masaki MORISHITA, Takaaki SAKAI, Kenji OG ...
    1999 Volume 65 Issue 635 Pages 2237-2242
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Experiments have been conducted on turbulence induced vibration of a circular cylinder in water flow with high Reynolds number (3×105<Re<3×106). Based on the power spectral density of cylinder vibration which was measured at several Reynolds numbers, data of fluctuating force coefficients, Strouhal number and also correlation length were obtained. As a result, it becomes clear that the prediction method based on the random vibration theory introducing the correlation length has sufficient margin for actual turbulence-induced response.
    Download PDF (667K)
  • Atsushi OKAJIMA, Tatsuji MATSUMOTO, Shigeo KIMURA
    1999 Volume 65 Issue 635 Pages 2243-2250
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    The forces acting upon and the flow patterns around a rectangular cylinder with a cross-section of width/height ratio of 0.2∼3.0 submerged in oscillatory flow were studied by using a U-tube water tank. Measurements were conducted of both in-line and transverse forces on rectangular cylinders in a wide range of Keulegan-Carpenter number (KC) from 1 to 160. The force signals were spectrally analyzed to determine the dominant frequencies, and the flow visualization technique was also employed to understand the correspondence between the flow patterns and the force coefficients. In this paper we discussed the influence of the width/height ratio of the cylinder on the flow patterns and the aerodynamic characteristics of a rectangular cylinder in oscillatory flow. We found that a drag coefficient CD of the d/h=0.6 cylinder becomes the biggest beyond KC&ap;120, compared with other case of the d/h cylinder in oscillatory flow. The good correlations between CD and wake width were shown also. From the predominant frequency components of the fluctuating transverse forces, it was found that the values of Strouhal numbers reduced by a maximum velocity, are similar with the results in the case of a uniform flow for d/h≤1.0.
    Download PDF (1198K)
  • Dong Wei LI, Shigehiko KANEKO, Shinji HAYAMA
    1999 Volume 65 Issue 635 Pages 2251-2256
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    This study reports the stability of annular leakage-flow-induced vibrations. The pressure distribution of fluid between a fixed outer cylinder and a vibrating inner cylinder was obtained in the case of a translationally and rotationally coupled motion of the inner cylinder. The unsteady fluid force acting on the inner cylinder in the case of translational and rotational single degree-of-freedom vibrations was then expressed in terms proportional to the acceleration, velocity, and displacement. Then the critical flow rate (at which stability was lost) was determined for an annular leakage-flow-induced vibration. Finally, the stability was investigated theoretically. It is known that instability will occur in the case of a divergent passage, but the critical flow rate depends on the passage increment in a limited range: the eccentricity of the passage and the pressure loss factor at the inlet of the passage lower the stability.
    Download PDF (735K)
  • Deng-Wei LI, Shigehiko KANEKO, Shinji HAYAMA
    1999 Volume 65 Issue 635 Pages 2257-2262
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    In this study, the stability of annular leakage-flow-induced vibrations was investigated theoretically and experimentally for a translationally and rotationally coupled two degree-of-freedom system. The critical flow rate was both theoretically and experimentally obtained as a function of the passage increment ratio and the eccentricity of the passage. A good agreement between the theoretical and experimental results was obtatined. It was discovered both theoretically and from the experiments that instability will occur in the case of a divergent passage: the eccentricity of the passage lowers the stability of the systems.
    Download PDF (670K)
  • Souichi SAEKI, Haruki MADARAME, Koji OKAMOTO, Nobukazu TANAKA
    1999 Volume 65 Issue 635 Pages 2263-2269
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    A self-induced sloshing is excited by the flow itself without any other external force. In a rectangular tank having a horizontal plane jet, the 1st mode sloshing grew in a certain condition of the inlet velocity and the water level. The self-induced sloshing in the rectangular tank was simulated, using a two-dimensional numerical simulation code. The simulated results agreed qualitatively with the experimental ones. In this study, a growth mechanism of self-induced sloshing in the numerical simulation was evaluated. The interaction between the free surface motion and the unsteady flow was modeled simply. The oscillation energy (En) supplied for sloshing motion was calculated from the pressure and velocity distributions in the tank. The oscillation growth was related to the sign, magnitude and distribution of supplied oscillation energy (En). Therefore, the excitation of the simulated self-induced sloshing was quantitatively explained by the present mechanism. The self-induced sloshing was considered to be caused by the nonlinear interaction between the jet and the sloshing motion without inlet-outlet flow. It was clarified that the excitation of self-induced sloshing was dependent on the distribution of the oscillation energy along the jet.
    Download PDF (1098K)
  • Souichi SAEKI, Haruki MADARAME, Koji OKAMOTO
    1999 Volume 65 Issue 635 Pages 2270-2277
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    A self-induced sloshing is excited by flow without any other external force. In a rectangular tank having a horizontal plane jet, the 1 st or 2 nd mode sloshing grow in a certain condition of the inlet velocity and the water level. In this study, the experiment of this phenomenon was conducted and physical characteristics were examined in detail. The new sloshing conditions, i.e. "1st Stage Sloshing" and "2 nd Stage Sloshing", were discovered in a tank. The governing parameter of the sloshing growth, i.e. the modified Strouhal Number Sts, was derived approximately. When Sts was nearly either 1.0 (1 st Stage) or 2.0 (2 nd Stage), the sloshing excitation was arranged consistently. The dependency of the inlet velocity and tank geometry was revealed using Sty. A sloshing mode shift or a jet mode (stage) transition was evaluated to give rise to, with varying the inlet jet velocity. Consequently, a feedback mechanism was considered to be validated, which was closed by the jet fluctuation and the sloshing motion. The overall physical oscillation mechanism of the self induced sloshing was clarified.
    Download PDF (1049K)
  • Masanori BABA, Haruki MADARAME, Koji OKAMOTO
    1999 Volume 65 Issue 635 Pages 2278-2285
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    A self-induced oscillation of a free-surface caused by plunging jet was discovered. Plane water jet flowed into a thin rectangular tank horizontally and fell to the water in the tank. The jet plunged into the water and flowed out through the outlet at the bottom. At a certain water level and flow rate, the free surface began to oscillate and the sloshing motion of water in the tank was excited. The occurrence of the oscillation was dependent on the jet velocity at the plunging point. The oscillating frequency almost agreed with the natural frequency of the liquid in the tank. Oscillation mode shift was observed with varying the tank width. The mode shift occurred so that the sloshing frequency fell in a certain range. Under oscillating conditions, Strouhal number, which represents the ratio of the travelng time of jet disturbance from the plunging point to the outlet to sloshing period, was in the range of 0.2 to 0.5.
    Download PDF (958K)
  • Haruki MADARAME
    1999 Volume 65 Issue 635 Pages 2286-2292
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    A sloshing can be induced by a round jet impinging on the free surface of water in a cylindrical tank. In this study, a growth mechanism of the sloshing was investigated applying the theory developed to explain "jet-flutter" in a rectangular tank. The movement of the impinging point results in additional fluid mass being left on the surface, which does not balance with the momentum supplied by the jet. When the force caused by the imbalance acts in phase with the sloshing motion, it supplies energy to the sloshing. The theory could estimated the sloshing condition fairly well. The measured phase relation between the jet displacement and the pressure gradient across the jet agreed with the theory.
    Download PDF (894K)
  • Toshio NISHIZAWA, Hiroyuki TAKATA
    1999 Volume 65 Issue 635 Pages 2293-2300
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Stall flutter of a compressor cascade is studied by numerical analysis of the unsteady separated flow around the blades oscillating in a torsional mode. A numerical method in developed for the present study, in which used are a discrete vortex modeling to calculate the convection and the diffusion of the vorticity in the separated flow and a grid system deforming together with the blade oscillation. The present method is applied to the compressor cascade working at such an inlet flow angle where the rotating stall is also liable to occur. The numerical result shows that the shedding of strong vortices from the leading edge of the blades has dominant influences both on the flow behavior and on the flutter excitation of the oscillating blades, and that the occurrence of the stall flutter is closely connected with the propagation of the rotating stall.
    Download PDF (1032K)
  • Toshio NISHIZAWA, Hiroyuki TAKATA
    1999 Volume 65 Issue 635 Pages 2301-2308
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Stall flutter of a compressor cascade is studied by numerical analysis of the unsteady separated flow around the blades oscillating in a torsional mode. When the cascade operates at a stalling inlet flow angle, the unsteady flow behavior varies according to the relationship between the phase propagation velocity of the blade oscillation and the propagation velocity of rotating stall. The propagation of rotating stall is synchronized with or entrained by the blade motion and becomes very regular when the both propagation velocities are close to each other. The oscillating blades are subjected to intense flutter excitation when the synchronization occurs. On the contrary, the rotating stall propagates regularly regardless of the blade oscillation when the both propagation velocities are largely different from each other. The propagation of rotating stall becomes irregular under the influence of the blade oscillation when the difference of the propagation velocities is in between. The flutter excitation of the blades never occurs in the latter two flow conditions.
    Download PDF (1033K)
  • Toshio NISHIZAWA, Yasuhiko IIDA, Hiroyuki TAKATA
    1999 Volume 65 Issue 635 Pages 2309-2316
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Stall flutter of a compressor cascade is studied by experiment using a wind tunnel with a linear cascade of blades oscillating in a torsional mode. The time varying pressure distribution on the blade surface and the unsteady velocity fluctuation in the blade passages are measured with pressure transducers and hot wire velocimeters. The experimental result shows that the synchronization and the entrainment occur between the blade oscillation and the rotating stall, and that these phenomena have dominant influence both on the unsteady flow behavior and on the flutter excitation of the oscillating blades. The experimental result agrees very well in general with the numerical result performed in advance using a discrete vortex modeling, particularly concerning the relationship between the occurrence of the stall flutter and the propagation of the rotating stall.
    Download PDF (878K)
  • Takao KOBAYASHI, Hiroyuki TAKATA
    1999 Volume 65 Issue 635 Pages 2317-2324
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Aerodynamic characteristics of three-dimensional unstalled cascade oscillating in a uniform incompressible flow are predicted numerically. Unsteady formulation is based on a small perturbation approach with use of potential equations expressed on a moving computational grid. The governing equations are discretized by the finite element method. A parametric study on unsteady aerodynamic characteristics is conducted for the three-dimensional rotor cascade of a low-speed axial-flow compressor. Flutter characteristics of blades oscillating with first three eigenmodes, i.e., first bending mode, second bending mode, and first torsion mode, are shown, and the distribution of flutter excitation along the blade span is described.
    Download PDF (906K)
  • Yoshinori NII, Tetsurou GOTANDA
    1999 Volume 65 Issue 635 Pages 2325-2332
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    An unexpected strong noise with high frequency components has been observed during a test operation of wind turbine rotor blades which were produced to have a sharp trailing edge and smooth surface for acoustic noise experiments. Acoustic measurements to investigate this noise have been carried out under various operating conditions, and the results were compared with those of the previous work. It was found that the noise was composed of intense discrete tones whose frequencies were distributed in the range approximately predicted by Fink's model based on the T-S instability wave over a flat plate. The level of the noise depended on the electricity output of the test machine: The level was high at high power output, while the noise was generated intermittently at low power output. It was also found that noise was suppressed by attaching tripping tape on the pressure side of the blades. It was belived that T-S instability wave developed in a laminar boundary layer was responsible for the generation of the observed tones.
    Download PDF (898K)
  • Takehiro HIMENO, Toshinori WATANABE
    1999 Volume 65 Issue 635 Pages 2333-2340
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    For the establishment of fluid management technology under low gravity conditions, the investigation methods with computational fluid dynamics are strongly desired to give appropriate assessment of designed systems for space application. In the present study, a numerical analysis method with C-CUP scheme coupled with Level Set method and CSF model was developed to simulate incompressible and compressible two-phase flows including surface tension effect, adhesion phenomena and gravity effect. By the application of CIP scheme to the re-initialization and advection of the level set function, the shape of curved interface between gas and liquid phases could be distinctly captured, and the surface tension was adequately estimated as a body force. With the developed code, the flow fields in a liquid tank under low gravity conditions were successfully simulated.
    Download PDF (864K)
  • Naoki TSUCHIYA, Haruo YOSHIKI
    1999 Volume 65 Issue 635 Pages 2341-2348
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Three dimensional viscous flow analysis was performed in a radial turbine rotor in which blade-to-blade velocity distribution had been measured by LDV. Calculations were executed both cases "with clearance" and "without clearance". Tip clearance area was filled with another computational grid and multi-block analysis code was adopted. The result of "with clearance" calculation was compared with experimental data. Good agreement was found in tendency of velocity distribution. Secondary flow occurring in rotor passage was investigated. At the inlet part of a turbine, passage vortex rotating counter to the rotor was observed. The mechanism of its generation could be explained dividing the vortex into the flows along the wall surfaces. At the exit part of a turbine, two calculations were quite different in the figure of secondary flow. Leakage vortex was growing and moving away from suction surface. It was clear that secondary flow was greatly affected by the flow discharging from leakage vortex. It shouldn't be considered that leakage flow is a local phenomena and can be distinguished from the flow between blades.
    Download PDF (910K)
  • Shinsuke MOCHIZUKI, Hiroaki OZAKI, Hideo OSAKA
    1999 Volume 65 Issue 635 Pages 2349-2357
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    In a turbulent boundary layer distorted by spanwise gaps between roughness elements, the skew-induced and stress-induced production terms in the vorticity transport equation have been examined experimentally. The skew-induced production maintains both vortices of negative and positive vorticity generated behind the spanwise gaps, and the reorientation mechanism is recognized in the contribution mostly given by a component -∂W/∂x·∂U/∂y due to the mean streamlines curvature depending on size of spanwise gap. The Reynolds shear stress component -<uw>^-, which also contributes to maintenance of the streamwise vortex through the stress induced production term, is correlated with the velocity gradients, ∂U/∂z and ∂W/∂y, rather than the mean streamline curvature ∂W/∂x. Meaning of the classification for gap size, i.e. S-gap, M-gap, and L-gap, appears in features of cross-stream distributions of the stress induced production due to non-isotropy of Reynolds normal stresses.
    Download PDF (891K)
  • Yorinobu TOYA, Ikuo NAKAMURA
    1999 Volume 65 Issue 635 Pages 2358-2365
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    This study reports experimental results of the free surface flow of rotating fluid between concentric cylinders which the inner cylinder is rotating and the other one is fixed. The free surface flow was developed up to the wave motion with increasing Reynolds number. The wave motion is presumed to be influenced by the inertia force or centrifugal force. As experimental results, the photographs of surface flow were taken by the 35 mm camera. They showed that the wave motion appeared in the whole flow region and then the ring shaped wave formed in the outside region of rotating flow. The variation of the position of free surface was measured by a supersonic sensor and analyzed by FFT method. The phenomena of the transition from the stable state to the chaotic one were clarified.
    Download PDF (1292K)
  • Takeshi KATAOKA, Michihisa TSUTAHARA, Satoshi MIZUTANI
    1999 Volume 65 Issue 635 Pages 2366-2373
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    The time development of two-dimensional fluid motion induced by a line sink of a rotating linearly stratified Boussinesq fluid in a finite-depth reservoir is numerically studied. The flow is characterized by a Froude number, Fr and the ratio of the inertial frequency, f, to the buoyancy frequency, N. Following the initiation of the sink, the inertial gravity wave modes are generated to propagate upstream, which results in the narrowing of a withdrawal-layer thickness. Attention is focused on investigating the evolution of the mode propagation and the resultant flow patterns for various Fr and f/N. The numerical results clarify that the nature of the mode propagation has little dependence on the parameter Fr if the distance from the sink is larger than a critical one, Nd/f where d is the reservoir's depth. The withdrawal layer thickness for Fr=0 is linearly proportional to the distance from the sink and the nonlinear effect (Fr〓0) can be easily incorporated only by adding the term representing the thickness under no rigid rotation. Furthermore, profiles of the withdrawal layer thickness show strong time-dependence whose period is about 2π/f.
    Download PDF (858K)
  • Susumu KYOMEN
    1999 Volume 65 Issue 635 Pages 2374-2381
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Free convection in a heated horizontal tube of laminar flow is caused by buoyancy forces due to the temperature differences between the pipe wall and the fluid. However, it was shown experimentally in previous papers (1994, 1995) that free convection in both a steady flow and a pulsating flow occurs when the temperature differences between air in a room and water in an unheated pipe are large. Moreover, the above phenomenon was also shown in numerical analyses for both a steady flow and a pulsating flow by the present author and colleagues previously (1994, 1997). To investigate flow patterns of pulsating flows in this paper, numerical analysis is made in the case that an amplitude of an oscillating velocity is large in comparison with the previous paper (1997), and the distributions of velocity, temperature, streamline and isotherm are shown schematically. Furthermore, the profiles of velocity and temperature gradients at the wall under the influences of frequency are presented in this paper.
    Download PDF (792K)
  • Hitoshi SUGIYAMA, Mitsunobu AKIYAMA, Takashi ARAI, Yasunori SHINOHARA
    1999 Volume 65 Issue 635 Pages 2382-2390
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    A numerical analysis has been performed for developing turbulent flow of non-Newtonian fluid in eccentric annular passage. Several calculations have been carried out to examine the drag reduction with decreasing power index of power-law fluids and the reduction of skin friction factor with increasing eccentricity defined by the distance between the centers of the inner and outer pipes for Newtonian fluid. In numerical analysis, the power law model for non-Newtonian fluid and an algebraic Reynolds stress model for turbulent flow were adopted in order to predict precisely non-Newtonian flow behaviour. Boundary-fitted coordinate system was introduced as the method of coordinate transformation. The numerical results are compared with the experimental data involving stramwise velocity and fluctuating velocities in axial, radial and tangential directions. As a result of comparison with the experiment, it was found that the present method could predict well the streamwise meam-velocity in both fluids and reproduce the secondary flow of the second kind generated by anisotoropic turbulence. As for the comparison of fluctuating velocities, the agreement with experimental data is more satisfied for Newtonian fluid than for non-Newtonian fluid. The calculated results also suggest that the drag reduction with polymer solution is realized by the present method as well as the experiment and the phenomenon of decaying fluctuating velocity with decreasing power index is predicted.
    Download PDF (854K)
  • Satoshi OGATA, Keizo WATANABE
    1999 Volume 65 Issue 635 Pages 2391-2397
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    In order to grasp the factor in drag reduction for a rotating disk with highly water-repellent wall, the pressure distributions and the velocity profiles were measured. It was seen that the flow angle of the highly-water repellent wall disk flow increased comparing that of a smooth disk. Frictional moment of an enclosed rotating disk in Newtonian fluids was analyzed using the boundary condition which the wall shear stress is proportional to the slip velocity of fluid at the rotating disk wall. The analytical results were agree well with the experimental data of a rotating disk with the highly water-repellent wall.
    Download PDF (695K)
  • Masakuni MURAMATSU, Takashi NAGATANI
    1999 Volume 65 Issue 635 Pages 2398-2405
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Traffic jams are investigated numerically and analystically in the optimal velocity model on a single-line highway. The condition is found whether or not traffic jams occur when a car stops instantly. It is shown that traffic soliton appears at the threshold of occurrence of traffic jams. The Korteweg-de Vries (KdV) equation is derived from the optimal velocity model by the use of the nonlinear analysis. It is found that the traffic soliton appears only near the neutral stability point. The soliton obtained from the nonlinear analysis is consistent with that of the numerical simulation.
    Download PDF (711K)
  • Shigetoshi YAMAMOTO, Fumiyasu KURATANI, Takenori OGAWA
    1999 Volume 65 Issue 635 Pages 2406-2412
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    In order to improve the performance of a multiblade fan, the effects of three types of suction cones of the fan casing on the fan efficiency and noise are investigated experimentally. The first type of the suction cone is the insertion type, which is inserted into the inside of the fan impeller. The second type is the extrusion type, which extrudes outside from the casing surface. The third type is the combination type of two types. The results of those three types are compared with those of the commonly used suction cone. The followings are made clear: (1) The insertion type and the extrusion type are effective in improving the efficiency and reducing noise. (2) The optimal lengths of the insertion and the extrusion exist. (3) The combination type is more effective in improving the efficiency. (4) The combination type with the skewed cutoff of the fan casing shows the best effect.
    Download PDF (795K)
  • Hideshi ISHIDA, Hideo KIMOTO
    1999 Volume 65 Issue 635 Pages 2413-2419
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    In order to study the chaotic behavior of the vibrational thermal convection in a square enclosure, calculation method and the feature of the average Nusselt number with vibration frequency were precisely examained. In the computation the Prandtl number, the Rayleigh number and the vibration Grashof number were held constant of 0.71, 104 and 106, respectively. The angular frequency of vibration was changed in the range between 10 and 7 680. The results showed that the phoenomena could be predicted with the calculation method adopted in this paper and the change in the time-dependent characteristics of surface-average Nusselt number with the angular frequency of vibration could be analyzed well with power spectra. This change was characterized by the five regimes proposed by Fu et al. Moreover it was clarified that the region where the hysteresis phenomena were detected corresponded to the one where the variation of the surface-average Nusselt number was irregular and aperiodic.
    Download PDF (975K)
  • Hideshi ISHIDA, Hideo KIMOTO
    1999 Volume 65 Issue 635 Pages 2420-2425
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    In this paper the time dependent characteristics of surface-average Nusselt number in a square enclosure with hot and cold side walls exposed vertical vibration were numerically examained. In the computation the Prandtl number, the Rayleigh number and the vibration Grashof number were held constant of 0.71, 104 and 106, respectively. The angular frequency of vibration was changed in the range between 10 and 7 680. The results showed that the change in the characteristics of the surface-average Nusselt number proposed by Fu et al. coresponded to the change in the shape of reconstructed attractor and that these regions could be also characterized by the three indices defined on the phase space: average location of trajectory, the largest Lyapunov exponent and correlation dimension. Moreover it was also revealed that the time scale with which the auto-correlated coefficient of the surface-average Nusselt number becomes 1/e was found to be very important parameter for time and surface-average Nusselt number.
    Download PDF (909K)
  • Toshiyuki MIYANAGA, Yukio NAKANO
    1999 Volume 65 Issue 635 Pages 2426-2433
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    A numerical calculation method of radiation heat transfer in a three-dimensional closed space including diffuse and specular surfaces has been developed. In order to enable to analyze the radiation heat exchange on each surface considering multiple specularly reflections and obstacles of radiation in the space, the improved heat ray tracing method was presented for calculating above-mentioned view factors accurately. It was examined to be applicable to complicated problems by test calculations using a few models with specular surfaces and obstacles. Continuing effort is applied to a more realistic problem. The example cited in this paper is the infrared emitter with a parabolic specular reflector, which is the common electric heating apparatus. The measured result of a radiant irradiated power distribution agrees well with the calculated one. The practical validity of our method was verified.
    Download PDF (1155K)
  • Hiroshi KAWANABE, Masahiro SHIOJI, Makoto IKEGAMI
    1999 Volume 65 Issue 635 Pages 2434-2438
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    A computational fluid-dynamic simulation for a turbulent nonuniform combustion is established using a stochastic approach. At each point in a turbulent flow field, the variations of species mass fraction and temperature are statistically described by the joint-probability density function, and the velocity variation is expressed using the conventional k-ε turbulence model. The transport equation of this joint pdf of mass fraction and temperature is calculated by a finite difference method in convection and turbulent diffusion and by the Curl collision-redispresion model in molecular mixing. This method is applied to simulate the process of scalar dispresion in a uniform isotropic turbulent flow. The results show that the profile of averaged scalar is quite similar to these calculated using conventional transport equation. Furthermore, a reasonable degree of reproduction is achieved for the pdfs of scalar at each point in the flow field.
    Download PDF (630K)
  • Masafumi HIROTA, Hideomi FUJITA, Ahmad SYUHADA, Shinji ARAKI, Makoto Y ...
    1999 Volume 65 Issue 635 Pages 2439-2446
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Heat/mass tranfer characteristics in rectangular cross-sectioned serpentine ducts with a sharp 180-degree turn have been measured by the naphthalene sublimation method, directing special attention to the influence of the flow-inlet condition on the local mass transfer. The test duct has relatively short straight sections, (7.9∼9.1) dh, before and after the turn. Two contrasrive flow-inlet conditions, contracted-flow inlet and uniform-flow inlet, have been tested for three turn clearances at Reynolds number of 3.5×104. The distributions of local Sherwood numbers on all walls of the test ducts are shown in the form of 2-D maps. In the straight section before the turn, the influence of the flow-inlet condition on the local mass transfer characteristics appears very clearly. After the turn section, the local mass transfer is also influenced by the flow-inlet condition in a duct with a large turn clearance. As the turn clearance is decreased, however, the influence of the flow-inlet condition becomes weaker.
    Download PDF (1215K)
  • Kenji KATOH, Kwing-So CHOI, Tsuneo AZUMA
    1999 Volume 65 Issue 635 Pages 2447-2453
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    A theoretical study of a turbulent channel flow was conducted to investigate the relationship between the heattransfer enhancement and the drag increase over rough surfaces using the timespace averaged momentum and energy equations. The results confirmed that the efficiency of the heat transfer surface, (i.e., the ratio of the heat transfer increase to the drag increase) is in general less than unity when the molecular Prandtl number Pr of the working fluid is less than the turbulent Prandtl number Prt. When Pr is greater than Prt, on the other hand, it was shown that the efficiency can be greater than unity as long as the roughness height is within the transitional region. It is suggested that the breakdown of the Reynolds analogy being experimentally observed over the heated riblet surface seems to be due to the inhomogeneity of the heat flux over the riblet surface when the thermal boundary layer is not yet fully developed.
    Download PDF (749K)
  • Hideo INABA, Akihiko HORIBE, Koichi OZAKI, Kazuo EMOTO, Hiroyuki KAKIU ...
    1999 Volume 65 Issue 635 Pages 2454-2461
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Experiment has been performed of heat transfer characteristics of the middle temperature latent heat storage system of the direct-contact heat transfer by using m-E (meso-Erythritol, melting point of 119°C, latent heat of 375 kJ/kg) droplets as a latent heat storage material and silicone oil as a heat transfer medium. In the present study the liquid m-E was injected into the heat tranfer medium through a circular nozzle. The m-E droplets changed from liquid to solid phase during falling in the heat transfer medium at low temperature. From the measuring results of m-E droplet diameter, falling velocity, and solidification rate, the nondimensional empirical equations of the arithmetic mean diameter of the droplets and falling velocity, the solidification rate and the overall heat transfer coefficient were derived as a function of the characteristic arithmetic mean diameter, the terminal velocity, temperature and physical properties.
    Download PDF (939K)
  • Masao NAKAZATOMI, Hideo SHIMIZU, Yasuhide KAKUNO, Koji MORI
    1999 Volume 65 Issue 635 Pages 2462-2469
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    In order to clarify the pressure effects on frictional pressure gradients in vertical upward gas-liquid two-phase flow, measurements of pressure gradients were made for a pressure range of 0.3∼20 MPa. Frictional pressure gradients for each system pressure were presented and it was revealed that the effects of superficial liquid velocity on the frictional pressure gradient gets smaller with increasing system pressure. Furthermore, the characteristic is clarified that there exist minimum values with increasing superficial gas velocity in the pressure range of 10∼20 MPa. Comparison of measured values of frictional pressure gradients with the correlations proposed by Lockhart-Martinelli and Chisholm were made and these correlations were found not to be valid for the data at high pressure conditions. A new correlation to predict friction pressure gradients was proposed.
    Download PDF (764K)
  • Hirohiko FUKUMOTO, Yoshinori KUROKAWA, Kazuhiko SAKIYAMA, Shigeto ADAC ...
    1999 Volume 65 Issue 635 Pages 2470-2475
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    The authors are doing research to develop a rapid curing device by using inductive heating for the production of a laminated rubber bearing. In the previous paper, it was shown that three problems exist; i.e. (i) the slit mold necessary, (ii) lower uniform temperature in the radial direction, and (iii) necessary temperature control by the internal location temperature, when the commericial frequency (60 Hz) induction heating was used to vulcanize a laminated rubber. This paper demonstrates that ultra low frequency heating is adequate to cope with the above problems. It will be shown that (i) the mold is not necessary a slit, (ii) the radial temperature difference is within ±5°C, and (iii) the internal temperature can be controlled by the side mold temperature, as the result of ultra low frequency induction heating test.
    Download PDF (708K)
  • Hirohiko FUKUMOTO, Yoshinori KUROKAWA, Yuji OIKAWA
    1999 Volume 65 Issue 635 Pages 2476-2481
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    The authors are studying to develop a rapid curing device by using inductive heating for the production of a laminated rubber bearing. As earlier paper reported on the possibility of rapid and high performance curing by using ultra low frequency heating. This paper reports on a test to manufacture φ500 laminated rubber using ultra low frequency heating. The test results certified that the time required for curing was decreased by 1/3 and the energy cost was decreased by 1/2, compared with tests performed using conventional hot wall heating. Finally, validity of practical rapid vulcanization using ultra low frequency induction heating was verified by a proof test using laminated rubber of mass production size (φ900).
    Download PDF (698K)
  • Tetsuo HIRATA, Mitsutoshi KATOH, Kouji NAGASAKA, Masaaki ISHIKAWA
    1999 Volume 65 Issue 635 Pages 2482-2488
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Experimental studies for freezing phenomena of ethylene glycol solution on a cooled horizontal plate have been performed. As the cooled plate, a glass, an acrylic resin, a polyvinyl chloride, a silicone and a copper plates are used. It is found that the crystal ice formed at some of the cooled plates is removed from the plate surface due to buoyancy force acting the crystal ice. It means that ice formation on a cooled plate without deposit ice layer is possible by the present method. It is shown that the cooled plate surface is undercooled about -2.0 degree below the freezing temperature of the solution during the crystal ice formation and its removal phenomena. The onset conditions for the ice removal phenomena are given by cooling heat flux and by thermal properties of the cooled plate and the solution.
    Download PDF (960K)
  • Shoji TAKADA, Kunihiko SUZUKI, Yoshiyuki INAGAKI, Yukio SUDO
    1999 Volume 65 Issue 635 Pages 2489-2497
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    An experiment was performed to simulate a water cooling panel system for passive decay heat removal from, a high-temperature gas-cooled reactor, comprised of a pressure vessel topped by a complex three-dimensional structure of stand pipes, to investigate the performance of decay heat removal and the temperature distributions of system components. The experimental apparatus consists of a pressure vessel of 1 m in diameter and 3 m in height, bristling with nineteen stand pipes atop and containing internal heaters with a maximum heating rate of 100 kW to simulate the decay heat of the core. The pressure vessel is surrounded by the water cooling panels. The analytical code THANPACST2 was applied to validate its analytical methods and a newly-proposed axi-symmetrical model simulating the stand pipes by porous body cells against the experimental data. Under the conditions of helium gas at pressure of 0.47 MPa and temperatures up to 430°C in the pressure vessel, the temperatures of the critical spots, where peak temperatures appeared near the top of the pressure vessel, were estimated within difference between -25 and +70°C from the experimental data. The heat transferred to the cooling panel was estimated to be 4.1% less than the experimental value.
    Download PDF (985K)
  • Seizo KATO, Yoshihiro YAMAMOTO
    1999 Volume 65 Issue 635 Pages 2498-2505
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    The paper describes quantum chemistry analysis using an ab initio molecular orbital method to know the microscopic mechanisms of plasma assisted reactions from methane into methanol. Two key successive reactions of CH4+O→CH3+OH and CH3+OH→CH3OH are dealt with and their geometrical configurations from the reactant to product via transition state, intrinsic reaction coordinates, reaction rate constants, etc. are calculated and discussed. The results make well clear the microscopic mechanisms of the two plasma reactions. The present quantum chemical approach is found to be a powerful tool to understand microscopic plasma reaction processes.
    Download PDF (875K)
  • Daisuke SEGAWA, Mitsuhiro TSUE, Toshikazu KADOTA, Hiroshi YAMASAKI, Mi ...
    1999 Volume 65 Issue 635 Pages 2506-2512
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Sooting field in a flame formed around a suspended droplet was visualized under microgravity using the planar laser light scattering technique. The soot concentration and the instantaneous soot amount were approximately estimated from the intensity of the scattered light using the image analysis system. The fuels employed were water in oil emulsions composed of n-dodecane, water and surfactant. The water content was varied from 0 to 0.2 in volume. A soot layer which was concentric with the droplet was observed inside the luminous flame. The results show the unsteadiness of the sooting behavior as well as the flame behavior. The maximum of the soot concentration is located near the inner edge of the soot layer. The time history of the instantaneous soot amount is similar to that of the instantaneous flame radius. The maximum of the soot concentration does not vary with the water content, while the soot amount decreases significantly with the water emulsification.
    Download PDF (932K)
  • Jiongming RUAN, Hideaki KOBAYASHI, Takashi NIIOKA
    1999 Volume 65 Issue 635 Pages 2513-2518
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Experimental and numerical investigations were performed for the Iaminar burning velocity and the flame structure of laminar premixed CH4/O2/CO2 flames. Measurements of the laminar burning velocity were conducted by using a flame cone angle method for a circular nozzle burner. Numerical simulation was performed using one-dimensional plane flame code including radiation heat loss with an optically thin model. It was shown that the laminar burning velocity decrease with CO2 addition even though the adiabatic flame temperature is the same as that for CH4/Air flames. The radiation heat loss is significant for the CH4/O2/CO2 flames, and the flame temperature and laminar burning velocity decreases when the radiation heat loss is considered. Effects of thermal properties, radiation, and chemical reaction on the determination of the laminar burning velocity of CH4/O2/CO2 flames were discussed.
    Download PDF (746K)
  • Susumu NODA, Shuhei YAMAMOTO, Yoshiaki ONUMA
    1999 Volume 65 Issue 635 Pages 2519-2525
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    The flame base structure of lifted diffusion flames identified as the triple flame has been investigated numerically. As the structure related to the flamelet model or the flame sheet model of the turbulent diffusion flame, the detail is required, because the structure may control the downstream flame. Thus, the reacting flows with four types of inlet fuel concentration gradient in the expanded duct examined by Kioni et al. or Hirota et al. are calculated by using the HSMAC method. Results show that the decrease in the inlet fuel concentration gradient leads to the growth of both premixed flames, especially the rich premixed flame, and weakens the diffusion flame. That results in the convex geometry of the lean premixed flame. Moreover, it has been found that the flame base structure of lifted diffusion flames is modeled in the mixture fraction space; the unburnt region upstream of the flame tip, the triple flame structure region surrounded by both premixed flames, the unburnt transient region outside of the triple flame region, and the diffusion flame structure region downstream of both of the triple flame region and the unburnt transient region.
    Download PDF (852K)
  • Akio KITAJIMA, Toshihisa UEDA, Akiko MATSUO, Masahiko MIZOMOTO
    1999 Volume 65 Issue 635 Pages 2526-2533
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    The flow field of the turbulent counterflowing stream with and without nonpremixed flames, and the behavior of diffusion region of turbulent nonpremixed flames have been investigated experimentally. The turbulent characteristics of each side flow in a counterflowing stream would be regarded as keeping their initial turbulent characteristics from one burner exit to near the stagnation region. And the mutual interferences of turbulence of both flows are slight in the present nonreactive turbulent counterflow field. The trends of velocity fields at every turbulent condition agree well with the flow field of nonreactive turbulent counterflowing stream and it is considered that the mutual interferences of both air side stream turbulence and fuel side stream turbulence seem to be slight in a turbulent counterflowing stream with nonpremixed flames. The behavior of diffusion region including the nonpremixed flame is mostly affected by the turbulence of the air stream. And it indicate that the presence of a certain close relation between the macroscopic behavior of diffusion region caused by the turbulent effect and extinction phenomena of turbulent nonpremixed flames formed in a counterflowing stream.
    Download PDF (864K)
  • Takashi SUZUKI, Hidenori SAWADA, Yasufumi OGURI, Masatake YOSHIDA
    1999 Volume 65 Issue 635 Pages 2534-2539
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    Although the relationship between luminous intensity of radicals and equivalence ratio φ has been investigated, the previous works underestimates an influence of high combustion pressure. The experiment was conducted by applying different initial charge pressure in the constant volume chamber and by changing φ. The result obtained indicates clearly that the luminous intensity for both CH, C2, and OH radical changes increases exponentially as the combustion pressure increases regardless of φ. Therefore each radical intensity can be approximated as a function of combustion pressure. It shows that luminous intensity of C2 radical is strongly influenced by the value of φ than that of CH, OH radical, especially in lean mixture state. It is verified that the measurement of φ can be accomplished using the ratio of luminous intensity of CH and C2 radical, and combustion pressure. The value of φ was estimated and compared with the actual value; the accuracy was about ±5%.
    Download PDF (607K)
  • Keiji KONISHI, Yasuo OKAMOTO, Haruo YOSHIKI, Shinichi TASHIRO
    1999 Volume 65 Issue 635 Pages 2540-2546
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    This paper describes some important results obtained through experiments on a small inward radial gas turbine driven by high temperature pulsating flows. The temperature ranges from atmospheric temperature to 800 K. The waveforms of pulsation were nearly sinusoidal and had various pressure amplitude but fixed frequency (50 Hz). A newly designed reduction device, which consists of planetary gears, helped the turbine to work as nearly as in realistic circumstances. Main results are as follows. Even under the high temperature conditions, the turbine characteristics, such as the mass flow rate and the power output, are the same as those under the low temperature pulsating flow conditions as far as the estimations are performed by using ordinary corrected values taking the temperature into consideration. As to the effect of pulsation, the more the amplitude of pressure increase, the less the mass flow rate through the turbine became at the same expansion ratio. Adding these results, the windage states characteristics of the turbine were made clear through experiments.
    Download PDF (809K)
  • Shoichi IWAMOTO, Koichi HIRATA, Fujio TODA
    1999 Volume 65 Issue 635 Pages 2547-2554
    Published: July 25, 1999
    Released on J-STAGE: March 28, 2008
    JOURNAL FREE ACCESS
    The authors have been manufactured and developed five kinds of high-and low temperature difference type Stirling engines. Their engine performance was investigated by experimental operations. And in order to consider about parameters affected to the engine performance, the experimental results were discussed and compared with the calculated results by analysis simulation methods. This paper shows an arrangement method for the experimental results, and considers about the performance of general Stirling engines. After using the arrangement method with non-dimensional numbers by a dimensional analysis, a prediction metiod, which is used in an early design process, is lead. One of the non-dimensional numbers in this prediction method is calculated from the engine specifications including specifications of the working gas. Then, the prediction method has the characteristics that are able to predict the engine speed, the output power, effects of kinds of the working gas and the operating conditions.
    Download PDF (940K)
feedback
Top