JSME International Journal Series B Fluids and Thermal Engineering
Online ISSN : 1347-5371
Print ISSN : 1340-8054
ISSN-L : 1340-8054
Volume 39 , Issue 4
Showing 1-25 articles out of 25 articles from the selected issue
  • Yasuhiko Sakai, Ikuo Nakamura, Hiroyuki Tsunoda, Kouta Hanabusa
    1996 Volume 39 Issue 4 Pages 667-675
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    A generalized Langevin equation developed by Haworth and Pope [Phys. Fluids, 29-2 (1986), 387] is applied to calculate the dispersion of a passive contaminant in turbulent pipe flow. The model coefficients in the equation are determined from an algebraic relation based on the consistency condition for the second-order moments of velocity, which includes the third-order moments. In the present model, the first-and second-order moments are used as input data, but the third-order moments are not inputted due to lack of reliable data. First, we confirmed numerically the consistency condition for a simulated velocity field. Second, the long-time diffusion was examined. The Eulerian velocity statistics show good agreement with the prescribed data up to second order. With regard to long-time diffusion, the longitudinal distributions of a cross-sectional mean concentration agree well with experiments. It is also found that the appropriate value for the Kolmogorov constant Co is 1.9 for this problem.
    Download PDF (1064K)
  • Mamoru Tanahashi, Toshio Miyauchi
    1996 Volume 39 Issue 4 Pages 676-684
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    Assessment of subgrid-scale (SGS) models in large eddy simulation was conducted using direct numerical simulation data base of a temporally developing turbulent mixing layer. The direct numerical simulation data base carefully calculated by means of a spectral method was used for comparison of SGS models. In this paper, various SGS models for incompressible turbulence were tested to clarify the characteristics of the models. The Smagorinsky model, the Bardina model and the GNS model predict the mean value of Reynolds stress very well, whereas the local structure of the Reynolds stress cannot be well represented by these models. Among the three models, the Bardina model predicts the local structure best. As for the Cross terms, the Bardina model shows high correlation with DNS data. The dynamic eddy viscosity model for total SGS stress predicts very small stress which is close to the Reynolds stress.
    Download PDF (944K)
  • Toshio Miyauchi, Mamoru Tanahashi, Kunihiko Miyake
    1996 Volume 39 Issue 4 Pages 685-691
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    Direct numerical simulation of three-dimensional compressible homogeneous isotropic turbulence was conducted by a spectral method to verify SGS models of large eddy simulation. The results obtained using a linear combination model and dynamic eddy viscosity model were compared with the exact SGS stress evaluated from the DNS data. The following conclusions are obtained. (1) Correlation coefficients between the exact cross term and the one predicted using the linear combination model and the original scale similarity hypothesis show values as high as 90%. (2) The SGS Reynolds stress predicted using the assumption of scale similarity shows the highest correlation with the filtered DNS data, while the correlation between the exact value and the value predicted using the linear combination model is low. (3) The stress predicted using the dynamic SGS model and the exact stress show low correlation which depends on the filter selection and the averaging volume.
    Download PDF (830K)
  • Hiroshi Maekawa, Akihiko Hirata, Kiyoshi Yamamoto
    1996 Volume 39 Issue 4 Pages 692-697
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    A numerical scheme has been developed for solving the two-dimensional incompressible Navier-Stokes equations on a domain that is infinite in the vertical (y) direction and finite in the streamwise (x) direction. The fourth-order equation for the streamwise velocity (u) is advanced in time explicitly using a compact third-order Runge-Kutta scheme. A standard Fourier method is used in the x direction and a mapped spectral method in the y direction. The various parts of the code are tested by solving two problems with analytical or independently-established known solutions. The nonlinear Stuart solution is employed to test the convective parts, the Poisson part of the code and the time advancement. Finally, we find that the disturbance obtained from linear stability theory grows in the linear regime of a free shear flow and that the nonlinear growth of the amplitude corresponds to the vortex roll-up.
    Download PDF (716K)
  • Hiroshi Ikeda, Takahiko Tanahashi
    1996 Volume 39 Issue 4 Pages 698-705
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    We have proposed a new hybrid-streamline-upwind finite-element method, which is based on the finite analytic method developed in the field of the finite difference method, and verified that this method is valid at high Reynolds numbers for unsteady problems using a one-dimensional Burgers equation. In this paper, we will show that this method is valid for steady two-dimensional advection-diffusion problems. Examples demonstrate the effectiveness of this streamline-upwind method for a dual space, excluding both wiggles and spurious crosswind diffusion which afflict some classical upwind schemes. Furthermore we achieve a simple formulation and high-speed calculation using the finite-element method, by combining the advection term and the diffusion term.
    Download PDF (778K)
  • Takanori Take, Teruhiko Kida, Tomoya Nakajima
    1996 Volume 39 Issue 4 Pages 706-713
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    The panel method in which vorticity is distributed on the surface of a body, the surface vorticity distribution method such like the vortex lattice method, is useful for determining the Euler flow past bodies and recently has been applied to complicated body shapes together with the vortex method. In the present paper we demonstrate the mathematical analysis of two panel methods for inviscid incompressible flow past two-dimensional bluff bodies, the pointwise and piecewise-linear distributions of vortices. The governing singular integral equation is derived by distributing vortices on the surface of the body and the simultaneous linear algebraic equations are derived by approximating the integral equation by the discretization of vorticity distribution. The analytic solution of the linear algebraic equations is obtained and the accuracy of the approximate solution is discussed. We show that : (1) there exists an appropriate collocation point, (2) the accuracy of the vortices on the surface of the body is of the order of 1/n where n is the panel number and (3) the eigensolution of the governing integral equation cannot be obtained using these schemes unless they are modified.
    Download PDF (822K)
  • Masaaki Tamagawa, Teruaki Akamatsu
    1996 Volume 39 Issue 4 Pages 714-720
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    This paper describes pressure wave analysis in order to examine the effects of shock waves on living tissues. From the comparison of displacements of living tissue models by FEM with that obtained by experiments, dynamic properties of living tissue (rat's kidney) are determined. Using these properties of the living tissues in shock compression (solid-state equation), the flow fields in two layers that consist of air and the living tissue, and in three layers that consist of air, water, and the living tissue are numerically obtained. It is found that the incident time derivative of pressure in the living tissue is smaller than that of pressure in air and water. Hence, the effect of the living tissue depends on the relaxation of pressure waves and duration of exposure.
    Download PDF (783K)
  • Kazutaka Kitagawa, Tetsuya Yasue, Seiji Watanabe, Tadamasa Jyounouchi, ...
    1996 Volume 39 Issue 4 Pages 721-730
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    One-dimensional interaction between a planar shock wave and porous material with various porosity φg are investigated numerically. As the result of numerical investigations, it is deduced that the dynamic behavior of gas-foam interactions strongly depends on the porosity φg of the foam. Four kinds of porous material are treated : rubber which is of no porosity (φg=0); foam 350×70×70 which is of low-porosity (φg=0.76) and high density (ρc=290kg/·); foam 50×50×50 which is of high-porosity (φg=0.98) and low density (ρc=26kg/·); and foam 13×13×13, which has the same density and porosity as foam 50×50×50, but has a different internal structure of foam material. In the no porosity (φg=0) and low porosity (φg&ap0∼0.76) cases, comparisons of numerical results with experimental ones show that the mobility of gas inside the materials is very low, and its dynamics can be approximated by a single-phase elastic motion. Results of single-phase analysis are compared with the shock tube experiments, which show essentially same stress variations with experimental results. In the high-porosity (φg&ap0.95) case, the mobility of the gas inside the foam is higher, and the dynamics must be treated as a two-phase flow. Results of two-phase analysis simulate experimental stress variations in high-porosity foam well, while the single-phase analysis for high-porosity foam dose not simulate experimental results well.
    Download PDF (970K)
  • Shinichi Yuu, Toshihiko Umekage
    1996 Volume 39 Issue 4 Pages 731-739
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    Turbulence causes collisions between neighboring particles by random motions. These collisions then result in coagulation of particles. In this study, we measured the change of size distributions of the suspended aerosol particles in a stirred tank using a scanning electron microscope. We also calculated the process of turbulent coagulation using a population balance equation with three different turbulent collision rate equations. The calculated results based on Saffman and Turner's equation which had neglected the relative velocity between colliding particles due to their inertia underestimated changes of size distributions. The calculated results based on Abrahamson's equation which had neglected the velocity correlation between colliding particles overestimated them. On the other hand, the calculated results based on Yuu's equation which had taken into account both the relative velocity due to their inertia and the velocity correlation between colliding particles were found to be in good agreement with the experimental results. It is concluded that the particle inertia and the correlation between collision particles should not be neglected for the description of particle collisions in turbulence even when the particle sizes are less than one micron.
    Download PDF (1066K)
  • Yoshinori Kita, Hideo Takase, Teruyuki Tokunaga, Ichiroo Kondoo, Shinp ...
    1996 Volume 39 Issue 4 Pages 740-746
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    Hydrous electrorheological (ER) fluid was poured into the annular interstitial space between two coaxial cylinders : the outer cylinder can rotate on its axis and the inner stationary one is connected to a torque meter. Upon application of an electric field and rotation of the outer cylinder, measurements of velocity distributions of ER fluid in the annular space were made using a laser Doppler velocimeter and those of shear stress exerted on the inner cylinder surface were made using the torque meter. When the rotational speed was very low, the flow was observed directly. Since high-water-content ER fluid requires a high electric current, a countermeasure for reducing the current was also examined. It was found that the ratio of electrical force to viscous force (= the inverse of the Mason number) plays an important role in determining the flow pattern of ER fluid and, upon increasing the magnitude of the ratio, the flow pattern varies from Couette flow to plug flow. Insulation of the inner cylinder surface with thin polyethylene films and a rubber membrane reduced the electric current but the magnitude of the generated shear stress decreased to one-tenth of that without insulation films.
    Download PDF (787K)
  • Tatsushi Mori, Tadashi Yanagisawa, Takashi Shimizu, Hiroki Tagami
    1996 Volume 39 Issue 4 Pages 747-753
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    The occurrence of periodic self-rotation of the orbiting scroll in a scroll compressor having an orbital mechanism composed of a ball coupling and an eccentric bushing was reported in our previous paper. In the present study we investigated the self-rotational motion of the orbiting scroll in detail. Complex motion of the orbiting scroll and the eccentric bushing was analyzed theoretically based on equations of motion. The motion was also measured experimentally using an optical method, at various operating pressures and magnitudes of clearance at the ball coupling of the compressor. Positive and negative self-rotation of the orbiting scroll during one revolution of the compressor was clearly demonstrated, and the experimental results agreed well with the results of the theoretical analysis.
    Download PDF (758K)
  • Yukio Yamada, Yasuo Hasegawa
    1996 Volume 39 Issue 4 Pages 754-761
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    Time-resolved spectroscopy is expected to be developed for optical imaging of biological tissues which strongly scatter and weakly absorb light. It is necessary to understand the transmission behavior of light impulses incident on such scattering and absorbing media. One of the numerical methods used to solve the photon diffusion equation is the finite element method (FEM). We report the FEM results of temporal variation of transmittance of light impulses through scattering slabs and cylinders. FEM results of transmittances through homogeneously scattering and absorbing slabs are shown and compared with the results obtained by a Monte Carlo method which is believed to provide more accurate results, although much longer computation time is necessary for a Monte Carlo method. The FEM results for cylinders with homogeneous scattering and inhomogeneous absorption are also presented. The effects of the location and size of the localized absorbing volume on the transmittances are investigated.
    Download PDF (936K)
  • Sei-ichi Iida, Kakuji Ogawara, Satoshi Furusawa
    1996 Volume 39 Issue 4 Pages 762-767
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    The Benard thermal convection field is known to have multiple solutions for a certain Rayleigh number condition. This means that several convection flow patterns can be stably sustained without an external control force. This study demonstrates that bimodal switching control is possible by means of flow state estimation. A neural network was used to recognize the transition state of the convection flow pattern. In order to change the flow pattern, an unstable feature of Benard convection was utilized. That is, the chaotic nature of wobbling around multiple solutions contributes to fast switching between stable modes. Numerical simulations were carried out to show the ability of the current control method.
    Download PDF (760K)
  • Masanori Monde, Yuhichi Mitsutake, Akikazu Kurihara, Shinichi Mihara
    1996 Volume 39 Issue 4 Pages 768-779
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    An analytical study has been done of critical heat flux of a two-phase thermosyphon, in which a liquid film and a vapor flow exist in a countercurrent annular flow. The CHF point on the thermosyphon is proved to correspond to a maximum falling liquid rate fed to the thermosyphon, which can be determined from three equations of momentum, its partial derivative with void fraction, and mass balance in the thermosyphon. This maximum point, furthermore, becomes identical to the point at which an envelope line generated from momentum equation and its partial derivative, intersects the mass balance line. The CHF calculated from the maximum liquid rate, is found to be in fairly good agreement with the existing CHF data in a closed thermosyphon and the CHF data measured for open themosyphon. Normal operation of the thermosyphon cannot be obtained at any liquid of water, R113, and R22 for a small tube diameter of less than 2 mm within a wide range of L/D-4.8 to 960 and density ratio of ρι/ρG=6.17 to 1602.
    Download PDF (1300K)
  • Takashi Suzuki, Koshi Mitachi, Hironori Yokoo
    1996 Volume 39 Issue 4 Pages 780-788
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    Heat transfer from a heated wire and a heated vertical plate, located just below a liquid surface, was studied experimentally. The curve representing the heat transfer coefficient as a function of the temperature difference between the heaters and a cooled tray could be divided into four parts. The range of each part depended strongly upon the size of the heaters, the depth of the tray and the liquid properties. The mechanism of heat transfer from the heaters in each part was discussed. The following observations were made. In the first part, where the temperature difference was the smallest, convective heat transfer was obscured by conduction. The heat transfer was mainly due to natural convection in the second part, and was mainly due to Marangoni convection in the fourth part. The third part could be considered as a transition regime. Furthermore, it was found that the transition was suppressed by the meniscus of the liquid surface which was in contact with the heaters.
    Download PDF (986K)
  • Kazunari Chikanawa, Suguru Yoshida, Haiping Hong, Hideaki Horio
    1996 Volume 39 Issue 4 Pages 789-797
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    Bubbles generated in conductive ink heated directly by electric current pulses can be used in an ink-jet printer to eject ink-droplets. Such parameters as the time required for bubble nucleation, growth rate of the bubble, maximum radius of the bubble and electric power input required for bubble generation, which depend strongly on the shape and size of electrodes immersed in the ink, influence the performance of the printer. A method of numerical analysis was developed to investigate the temperature distribution of the ink and dynamics of the bubble, and was confirmed to give values of the above parameters which were in good agreement with those of experiment. An optimum configuration of the electrodes was suggested, based on the comparison of results calculated for different shapes and dimensions of the electrodes.
    Download PDF (1103K)
  • Toshiharu Oka, Yoshiyuki Abe, Yasuhiko H. Mori, Akira Nagashima
    1996 Volume 39 Issue 4 Pages 798-807
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    Microgravity pool boiling experiments were carried out utilizing the drop shaft at the Japan Microgravity Center, which enables control of microgravity to within one-ten-thousandth of terrestrial gravity for a period of 10 seconds. Vapor bubbles were generated on a Joule-heated stainless steel or ITO film backed with a thin alumina or glass substrate, which was immersed in a pool of subcooled or saturated liquid at atmospheric pressure. The CHF (critical heat flux) to CFC-113 was lowered, in microgravity, to one-fifth of that achieved under the normal terrestrial condition. In contrast, the CHF to water remained more than one-half that achieved under the terrestrial condition. This difference in the gravity dependence of CHF is presumably ascribable to the difference in the frequency of bubble detachment from the heater surface between CFC-113 and water in microgravity, which is dependent on the difference in thermophysical properties, particularly the surface tension and the heat of vaporization, between the two fluids.
    Download PDF (1416K)
  • Kazuhiko Kudo, Akiyoshi Kuroda, Amr Eid, Tatuhiko Saito, Masahito Ogum ...
    1996 Volume 39 Issue 4 Pages 808-814
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    A method is developed for solving inverse radiative load problems within multidimensional systems with arbitrary shapes. The energy equations for each gas and wall element are expressed in a matrix form, and the inverse matrix is obtained by the singular value decomposition techniqe in order to estimate the temperature and heat load distributions within the gas region from the profiles of the heat flux and the temperature of the wall elements. From these analyses, the method is shown to give stable results for the radiative inverse load problems when the condition number of the matrix is reduced to an appropriate low value by setting some singular values to 0.
    Download PDF (889K)
  • Tamio Ida, Kazutomo Ohtake
    1996 Volume 39 Issue 4 Pages 815-821
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    The authors discuss in detail the various heat transportation mechanisms existing in turbulent diffusion flames by use of spectral analyses. Space and time-resolved structures of turbulent diffusion flames were analyzed by two-point laser Rayleigh spectroscopy (LRS) which did not directly interfere with the combustion media during measurement. The turbulent diffusion flame structures were divided into 4 regions based on the characteristics of their spectral analysis of time-dependent temperature signals. In order to discuss the macroscopic heat transportation mechanism, the coherent function from cross and power spectral functions at Region I∼IV was analyzed. From these analyses, the following diffusion characteristics at each region were revealed. Region I : three-dimensional diffusion mechanism (x-, r-, z-axes), Region II : one-dimensional diffusion mechanism (z-axis), Region III : two-dimensional diffusion mechanism (x-, z-axes), Region IV : no specific diffusion mechanism.
    Download PDF (790K)
  • Michio Kitano, Taro Kawamura
    1996 Volume 39 Issue 4 Pages 822-829
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    We investigated the effects of fuel mixing on flame height and flame temperature of coflow diffusion flames which were closely related to soot formation inside the flame. Four fuels with different sooting tendencies and transport properties (CH4, C3H8, C2H4, C2H2) were diluted with nitrogen and used as component fuel gases for mixing. The main results obtained are as follows : (1) Flame tip temperature is a good qualitative measure of the intensity of soot formation inside the flame. (2) Addition of nitrogen to the fuel causes both dilution and soot suppression, which lead to the decrease and the increase, respectively, in the flame tip temperature, and at a certain addition rate the flame tip temperature attains a maximum due to their competitive action. (3) There exists a strong correlation between the degree of soot formation and adiabatic flame temperature, and when soot formation is vigorous, factors which increase this temperature promote soot formation. (4) Effects of Lewis number and preferential diffusion should be taken into account to achieve effective suppression of soot formation by mixing fuels with different transport properties.
    Download PDF (951K)
  • Kaoru Furushima, Yukio Kawano, Kazuhiro Aoyama, Yoshiaki Onuma
    1996 Volume 39 Issue 4 Pages 830-835
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    A combustion model for turbulent diffusion flames is estimated frequently through the comparison of the simulated result with the experimental one. Usually, profiles of time-averaged concentration and temperature are used in that comparison, because the local reaction rate cannot be directly measured. However, since their profiles are also influenced largely by transport phenomena, it is difficult to estimate the combustion model properly with this method. Therefore, it is desirable to calculate the local reaction rate from experimental results and compare it with the simulated one. In the present study, from this point of view, the local reaction rate was tried to obtain by numerical calculation using measured values for a hydrogen jet diffusion flame. Then, it was suggested through the comparison of the obtained result with the simulated one that the method proposed here can provide reliable values for the local reaction rate.
    Download PDF (698K)
  • Yoshisuke Hamamoto, Eiji Tomita, Masaki Yamanaka, Motoshi Kataoka
    1996 Volume 39 Issue 4 Pages 836-843
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    It is necessary to know the heat flux near a wall in order to understand the unsteady heat transfer in a combustion chamber. In this study, a fuel-air mixture was ignited at the center of a constant-volume vessel and the heat flux was determined from the change in the temperature distribution in the boundary layer near the wall of the end gas side by analyzing the interference image. The heat flux was compared with that from a thin surface coaxial thermocouple with high response. It was found that before the flame arrives at the wall, the heat transfer is smaller than the thermal radiation from the flame, and the temperature of unburned gas changes nearly adiabatically. After the flame reaches the wall, the heat flux due to the temperature gradient becomes larger than that due to the thermal radiation.
    Download PDF (876K)
  • Michio Kitano, Jun Shibata
    1996 Volume 39 Issue 4 Pages 844-851
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    Premixed combustion of liquid fuel (kerosene) by air bubble blowing was attempted, which is a new method originated from the reverse concept of spray combustion. Setting a porous ceramic plate above the foam layer formed by blowing air into fuel made it possible to stabilize premixed flames on the plate. For most conditions examined the surface of this flame-holding plate was heated and became red-hot, which was effective for promoting fuel evaporation. The dependencies on air flow rate and equivalence ratio of temperature distributions inside and outside the plate were measured and examined in comparison with the cases for uniform gas mixtures. The burned gas was also analyzed in the downstream region and it was found that CO/CO2 level and UHC concentration were less than the regulation values over a wide range of combustion intensities up to the turn-down ratio of 4.8, and NOx concentration was nearly on the same level as other combustors for domestic use.
    Download PDF (917K)
  • Shouichi Ito, Katuo Asato, Takeshi Kawamura, Yoshihiro Ito
    1996 Volume 39 Issue 4 Pages 852-858
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    The effects of the temperature of the stagnation surface (surface temperature), the Lewis number of the deficient reactant and preheating of the unburned gas on the characteristics of premixed flat flames with heat loss stabilized in the stagnation flow were numerically examined in order to utilize the heat from the flame effectively and to develop a new type of domestic burners and heaters. Then, the characteristics of convective heat transfer to the stagnation surface from the flame were discussed. (1) With an increase in the surface temperature Tw and in the preheat temperature of the unburned gas Te, and with a decrease in the Lewis number Le, the flame temperature at the same value of the first Damkohler number DI is increased. (2) The heat flux utilized from the flame at the near limit of extinction is increased with an increase in Te, and with a decrease in Le and Tw. (3) The characteristics of heat transfer to the stagnation surface in the case in which the premixed flat flame is stabilized in the stagnation flow are superior to those in the stagnation flow in which only the burned gas flows. (4) In order to increase the heat flux utilized from the premixed flat flame stabilized in stagnation flow, it is necessary to maintain higher flame temperature and to stabilize the flame near the stagnation surface. Therefore, preheating of the unburned gas and use of the premixed gas with Le<1.0 are effectual in improving characteristics of heat transfer to the stagnation surface.
    Download PDF (839K)
  • Jiro Senda, Masaaki Kobayashi, Seiji Iwashita, Hajime Fujimoto
    1996 Volume 39 Issue 4 Pages 859-866
    Published: November 15, 1996
    Released: February 18, 2008
    JOURNALS FREE ACCESS
    We present a new model of the dispersion process of nonevaporating and evaporating diesel sprays impinging on d flat wall. The models concerning fuel film formation on the wall and its breakup process due to the droplet impingement, the breakup of an impinging droplet owing to boiling at the liquid-solid interface and the heat transfer from the wall to a droplet were constructed for both sprays. They were incorporated into KIVA original code. The calculated results obtained using the new models show good qualitative correspondence with the experimental results, compared with those obtained using KIVA original code.
    Download PDF (871K)
feedback
Top