Thermal Science and Engineering
Online ISSN : 1882-2592
Print ISSN : 0918-9963
ISSN-L : 0918-9963
Volume 15 , Issue 3
Showing 1-8 articles out of 8 articles from the selected issue
  • Takato SATO, Yasuo KOIZUMI, Hiroyasu OHTAKE
    2007 Volume 15 Issue 3 Pages 101-109
    Published: 2007
    Released: February 25, 2008
    JOURNALS FREE ACCESS
    Pool nucleate boiling heat transfer experiments were performed for water by using a well-controlled and -defined heat transfer surfaces in the range of the surface heat flux of ∼ 4.54×104 W⁄m2. One or three cavities were created on a mirror-finished silicon plate of 0.525 mm thickness by utilizing the Micro-Electro Mechanical Systems (MEMS) technology. In present experiments, the cavities were arranged in a straight line. The silicon plate was placed facing up at the bottom of the test container filled with distilled water. The back side of the silicon plate was irradiated by a laser beam to heat up the test heat transfer surface. The back side temperature was measured with a radiation thermometer. A boiling state was recorded with a high speed video camera. Thermal interaction between neighboring cavities became weak as the cavity spacing became wide and it disappeared when S⁄Lc = 1.6 in present experimental range. Four bubble coalescence patterns; vertical, horizontal and declining coalescence and vertical lift (no coalescence), were confirmed. When S⁄Lc ≥ 1.6, horizontal and declining coalescence disappeared. When the cavity spacing was narrow, hydraulic interaction between neighboring cavities played an important role in heat transfer. It became less important as the cavity spacing became wide. When S⁄Lc ≥ 1.2, the hydraulic interaction between neighboring cavities became negligible and phase change heat transfer took a main part.
    Download PDF (901K)
  • Tatiana ZOLOTOUKHINA, Toshihiro IWAKI
    2007 Volume 15 Issue 3 Pages 111-118
    Published: 2007
    Released: February 25, 2008
    JOURNALS FREE ACCESS
    The equation of wave propagating in fluid is described as a differential equation of the velocity or small element density from the Navier-Stokes (NS) equation. On the other hand, the molecular dynamics (MD) equation expresses the motion of a particle constituting the fluid and shows that the particle is always in motion regardless of the existence of the wave. In this study, we discuss in what way meanings of the velocity and density in the wave equation can be adopted in the MD system. What are the differentials with respect to time and space in the NS equation for the system of MD particles? Ordinarily, the physical quantities in the NS equation are obtained as the ensemble and time averages over the MD system. We investigate the number of particles and duration of time that are sufficient for the averages and simultaneously confirm whether the averaged values satisfy the differential equation. The two-dimensional MD method is used for the qualitative understanding. The fluid is assumed to consist of particles connected by the Lennard-Jones potential. The satisfaction of the differential equation by the MD averaged values is shown by the propagation velocity in the wave equation. The propagation velocity can be also obtained in another way i.e. by the ‘observation’ of the wave fronts motion of velocity, density or total energy wave in the fluid. The propagation velocity that has resulted from the wave equation is strongly affected by the ensemble and time averages. On the other hand, when it is obtained from the wave front, it is independent of the ensemble and time averages. We can have the former propagation velocity close to the latter one when a long time average is used for a large ensemble and a short time average for a small ensemble, i.e. the product of both averages can be considered changed at same rate as if by a scaling coefficient.
    Download PDF (1359K)
  • Masakazu HARASHIMA, Satoshi KAWAME, Hiroshi KAWAMURA
    2007 Volume 15 Issue 3 Pages 119-127
    Published: 2007
    Released: February 25, 2008
    JOURNALS FREE ACCESS
    Thermocapillary convection is driven by surface tension gradient due to temperature gradient along the free surface of a liquid. Thermocapillarity is of fundamental importance in material processing and in micro scale. A floating-zone method is a material process technique for producing and purifying single crystals of metals and oxides. In a half-zone liquid bridge, which mimics a half of a floating-zone method, the thermocapillary convection of a high Prandtl number fluid is induced by applying the temperature difference ΔT between cylindrical hot and cold rods sustaining a liquid bridge. If ΔT exceeds a critical value ΔTc, the flow field exhibits a transition from a two-dimensional steady flow to a three-dimensional time-dependent oscillatory one. The onset of the oscillation is known to be sensitive to heat transfer at free surface caused by the ambient air motion. Under the gravity environment, however, the thermocapillary convection is often hidden by the influence of the buoyancy. This problem is solved by experimental system with a small scale and with placing horizontal partition disks near both the top and bottom rods in the ambient air. With partition disk, the buoyant flow in the surrounding air can be suppressed and controlled. In the present study, an effect of the ambient temperature upon the stability of the thermocapillary convection is investigated experimentally and numerically considering the ambient region with or without partition disk. The transition ΔTc once increases and then decreases with increasing heat gain as pointed out by Kamotani (2001). It turned out that this change in ΔTc is accompanied by the transition of the azimuthal mode number.
    Download PDF (1509K)
  • Manabu OKURA, Kiyoaki ONO
    2007 Volume 15 Issue 3 Pages 129-139
    Published: 2007
    Released: February 25, 2008
    JOURNALS FREE ACCESS
    In order to keep environment in the air-conditioned room comfortable, it is important to predict air velocity and temperature field as precisely as possible. When walls or windows of a room cannot be assumed to be adiabatic, estimation of the heat transfer through them is one of the essential factors to perform satisfactory air-conditioning design of the room. Numerical simulation of the cooling process of a cubic shell is conducted. The Navier-Stokes equations governing flow field inside and outside the room and the heat conduction equation applying to walls are solved simultaneously, instead of assuming the heat transfer coefficient between the fluid and the solid surface. The computational result is compared with the experimental one. Their results show good agreement with each other; thus the validity of the calculation is confirmed. Characteristics of the heat flux and the heat transfer coefficient are discussed. It is found that the heat flux through sidewalls is larger than the ones through a top or a bottom wall. This is due to the thinner boundary layer at the lower part of sidewalls. The heat transfer coefficient shows the same tendency.
    Download PDF (2883K)
  • Hideo INABA, Fujio KOMATSU, Akihiko HORIBE, Naoto HARUKI, Akito MACHID ...
    2007 Volume 15 Issue 3 Pages 141-150
    Published: 2007
    Released: February 25, 2008
    JOURNALS FREE ACCESS
    This paper describes sorption characteristics of organic sorbent coated on heat transfer surface of a plate-fin-tube heat exchanger. The organic sorbent is a bridged complex of soldium polyacrylate. This bridged complex containing the carboxyl group as water vapor adsorption site has a larger adsorption abilities as compared with silica gel. The experiments in which the moist air was passed into the heat exchanger coated with sorption material were conducted under various conditions of air flow rate and the temperature of brine that was the heat transfer fluid to cool the air flow in the dehumidifying process. It is found that the sorption rate of vapor is affected by the air flow rate and the brine temperature. Meanwhile, the attempt of clarifying the sorption mechanism is also conducted. Finally the average mass transfer coefficient of the organic sorbent was non-dimensionalized as a function of Reynolds number and non-dimensional temperature. In addition, it was observed that the factor which affects the sorption rate in the water vapor sorption process of the organic sorbent coated on the heat exchanger shifts from the “adsorption step” to the “sorption step”.
    Download PDF (716K)
  • Takahiro TSUKAHARA, Kaoru IWAMOTO, Hiroshi KAWAMURA
    2007 Volume 15 Issue 3 Pages 151-162
    Published: 2007
    Released: February 25, 2008
    JOURNALS FREE ACCESS
    Direct numerical simulation of turbulent heat transfer in a plane Couette flow at Reynolds numbers of 3000 and 8600 (based on the relative wall speed and the channel width) has been performed with emphasis on a large-scale turbulence structure (LSS). Two kinds of temperature boundary conditions are employed; one is the uniform heat flux heating (UHF) and the other is the constant temperature difference between the walls (CTD). A series of DNS has been carried out for several larger computational box sizes than those of existing DNS's to examine their effect upon the LSS of the thermal field and the statistical quantities such as temperature variance. The differences caused by the boundary conditions of temperature field are discussed with respect to the LSS. For CTD, the LSS is remarkable in the outer region and is observed to be similar to that of the velocity field. Using the present largest box size of 96h×h×12.8h, the average streamwise and spanwise spacing of the LSS in both velocity and thermal fields are found to be λx = 32∼48h and λz = 2.1∼2.5h, respectively. In addition, it is concluded that the box needs to be larger than the size of the LSS in order to assess the accurate statistics such as Nusselt number.
    Download PDF (2786K)
  • Motoaki UTAMURA, Konstantin NIKITIN, Yasuyoshi KATO
    2007 Volume 15 Issue 3 Pages 163-173
    Published: 2007
    Released: February 25, 2008
    JOURNALS FREE ACCESS
    A generalized mean temperature difference (GMTD) method for heat exchangers is proposed. In the analysis of the performance of heat exchangers logarithmic temperature difference (LMTD) method has been widely used. This method, however, limits its application to those heating media with constant physical property. In turn GMTD method allows analysis with physical property distributed in an entire heat exchanger. Temperature profiles of the heat exchanger taken as function of heat load in place of axial position, mean temperature difference is evaluated numerically. It is mathematically demonstrated that LMTD method is an extremity of the GMTD method in the case of constant physical property. The GMTD method is applied to a hot water supplier with supercritical carbon dioxide as a heating media which is attracting attention as energy saving tactics. The hot water supplier operates under the condition of pseudo critical point of carbon dioxide where specific heat behaves anomaly. Incorporating GMTD method averaged overall heat transfer coefficient and subsequently formula of local Nusselt number are successfully derived for microchannel heat exchanger while formal application of LMTD method is found to give poor results i.e. two times less value with a larger error. This proves the validity of GMTD method.
    Download PDF (940K)
  • Takahiro ARAI, Masahiro FURUYA
    2007 Volume 15 Issue 3 Pages 91-100
    Published: 2007
    Released: February 25, 2008
    JOURNALS FREE ACCESS
    We proposed ultra rapid solidification and atomization technique, CANOPUS (Cooling and Atomizing based on NOble Process Utilizing Steam explosion), using small-scale vapor explosions to make an amorphous metal. The CANOPUS method is suitable for rapid cooling and atomization process, which utilizing sustainable small-scale vapor explosions. In order to apply the CANOPUS method to a high melting point metal, it is necessary to make a small-scale vapor explosion occur at a high temperature of the molten metal. Small-scale experiment is conducted to develop the vapor explosion promotor in which spontaneous vapor explosion can occur at a high temperature of a molten metal. Spontaneous vapor explosion do not occur when water at 80°C is used as a coolant. However, spontaneous vapor explosion occurs when water at 80°C with salt additives is used as a coolant. Specifically, lithium chloride solution generates spontaneous vapor explosions at the highest temperature of the molten tin in the experiment. In order to clarify the triggering mechanism of the spontaneous vapor explosion when the promotor is used as a coolant, a high-temperature solid stainless steel sphere is immersed into a coolant. The interfacial temperature of the stainless steel sphere is measured, and the behavior of a vapor film around the stainless steel sphere is observed with a digital video camera. As a result, salt additives resulted in an increase of quench temperature in all salt solutions. The quenching curves of each coolant indicate that the salt additives improve the film boiling heat transfer. The improvement of the film boiling heat transfer causes an unstable formation of the vapor film and a rise of the quench temperature. It is clarified that the salt additives to water promotes a vapor film collapse. Comparing two experiments, the quench temperature of each solution is in close agreement with the upper limit of the molten tin temperature that causes spontaneous vapor explosion. This result suggests that the vapor film collapse triggers spontaneous vapor explosion.
    Download PDF (2188K)
feedback
Top