The start-up flow of polymeric liquids between parallel plates was numerically simulated using a stochastic approach. The extra stress was calculated by solving the stochastic equation of a reptation model instead of using a constitutive equation. The numerical results showed that the present numerical simulation was capable to describe typical macroscopic behaviors of polymeric liquids such as the change in velocity profile with time, the growth in shear stress, and overshoot phenomena of velocity and wall shear stress. Moreover, the stochastic approach can predict microscopic behaviors such as the time development of polymer orientation. It is advantage of the present method that information related to the behavior of microscopic polymer structures in a flow can be obtained. In addition, the comparison with the numerical prediction using an approximated constitutive model showed that a closure approximation used led to a remarkable error just after the onset of the flow.
In this paper, the stability of a radial liquid sheet is described experimentally and theoretically. The liquid sheet was formed by the release of a radial liquid film flowing on a disk from the edge of the disk to still air. When the Reynolds number was large, a concentric disturbance wave (D wave) appeared and grew downstream on both free surfaces of the liquid sheet. To clarify the cause of the D wave, its frequency was measured by a laser beam reflection method and the cross section of the wavy liquid sheet was observed by the laser-induced fluorescence (LIF) method. Temporal and spatial linear stabilities of the liquid sheet were also investigated. The frequency and wavelength of the D wave and the phase difference between the D waves on the two liquid surfaces agreed well with those of the disturbance predicted using the stability theory. This means that the D wave on the liquid surface is caused by unstable disturbance, which is attributable to the inflectional velocity distribution of the liquid sheet. It was also found that Gaster's simplified equation, which transforms the temporal growth rate to a spatial one in the stability theory, is inadequate for a disturbance with a high growth rate.
The well known collapsible tube experiment was conducted to obtain flow, pressure and materials property data for steady state conditions. These were then used as the boundary conditions for a fully coupled fluid-structure interaction (FSI) model using a propriety computer code, LS-DYNA. The shape profiles for the tube were also recorded. In order to obtain similar collapse modes to the experiment, it was necessary to model the tube flat, and then inflate it into a circular profile, leaving residual stresses in the walls. The profile shape then agreed well with the experimental ones. Two departures from the physical properties were required to reduce computer time to an acceptable level. One of these was the lowering of the speed of sound by two orders of magnitude which, due to the low velocities involved, still left the mach number below 0.2. The other was to increase the thickness of the tube to prevent the numerical collapse of elements. A compensation for this was made by lowering the Young's modulus for the tube material. Overall the results are qualitatively good. They give an indication of the power of the current FSI algorithms and the need to combine experiment and computer models in order to maximise the information that can be extracted both in terms of quantity and quality.
Dust separation characteristics of a counter-flow vortex tube were investigated with lime powders whose mean particle sizes were 5 and 14.6µm. The experiment showed that a vortex tube can be used as an efficient pre-skimmer to separate particles from the waste gas in industry. In case of fine particles, a trend of increasing separation efficiency was observed with inlet pressure up to 160kPa at which the flow rate was 9.5m3/hr, but there was a tendency of reduction in the coarse particles. For both powder sizes, the efficiency and the effective nozzle inlet velocity were 93% and 14.52m/s respectively, while the tested vortex tube had a small diameter of 16mm and the ratio of the nozzle hole area to the tube cross-sectional area was 0.15. Additionally, the geometric ratios of the vortex tube could be proposed as 0.44 for the ratio of the orifice diameter at clean gas exit to the tube diameter and 14.0 for the ratio of the tube length to the diameter.
A computational model for flows in the plain-orifice atomizers with chamfered or rounded orifice inlets is established. The volume of fluid (VOF) method with finite volume formulation was employed to capture the liquid/gas interface. A continuum Surface Force (CSF) model was adopted to model the surface tension. The body-fitted coordinate system was used to facilitate the configuration of the atomizers. The evolution of the fluid/air interface and the velocity vector plots for the atomizers are discussed. The discharge coefficients and the spray angles for the atomizers are also compared. The result is explained by the profile of the axial velocity component at the atomizer exit and the evolution of the pressure drop. It is found that the discharge coefficient decreases very rapidly at the early stage of atomization, while the pressure drop has an abrupt rise at that time. With the same Reynolds number based on the orifice diameter and the mean axial velocity at the atomizer exit, the atomizer with a rounded orifice inlet has a larger discharge coefficient and a larger spray angle. For the conditions investigated in the present study, the atomizer with a rounded orifice inlet is beneficial for better atomization of the liquid jet.
This paper describes an experimental and numerical research on internal flow of a radial turbine impeller. The detailed flow patterns within the impeller were measured by LDV anemometer, and CFD code “BTOB3D”, which had been widely used in engineering, was used for flow simulation. The comparison of flow field was carefully and strictly conducted between computation and experiment. The results confirm that CFD computation can reach excellent qualitative agreement to experimental data, and most of flow characteristics can be captured by computation. The quantitative accuracy is also fine in large flow regions, however in some chord positions large error can be found for individual flow parameters. Finally, based on experimental data and computed results, the effects of incidence angle and flow distribution at impeller inlet are discussed.
Cavitation in a 3D axisymmetric rectangular groove occurred in various types of cavitation, such as vortex cavitation, cloud cavitation, traveling cavitation and sheet cavitation. In this experimental study, some aspects of cavitating flow were clarified. Prediction and experimental calculations were also made for incipient cavitation of both the cavitating vortex ring in the groove and the sheet cavitation at the downstream of the groove. The calculations indicate that the observed cavitation is not vaporous cavitation, but gaseous cavitation. The results of this study are useful for understanding the cavitation in this type of groove structure in order to control cavitation damage.
This paper describes a mathematical formulation and numerical implementation on the effect of surface impedance for reducing aerodynamic sound radiated from a circular cylinder. Firstly, mathematical derivation of the ideal surface impedance that eliminates the pressure fluctuation on the body surface, utilizing the Green's function, is discussed. Secondly, a numerical simulation is explained with using the Large Eddy Simulation (LES) techniques at Reynolds number 3000, with boundary conditions of, (i) rigid, (ii) finite uniform impedance, and (iii) our new approach of optimized surface impedance, inducing flow by a Helmholtz resonator. In the simulation only larger impedance than the ideal one is applied to the surface due to computational instability which restrains the lower limit of impedance. The effect of optimized surface impedance is verified by the simulation via the fact that the amplitude of the surface pressure fluctuation decreases remarkably.
A 1.829m (6ft) diameter industrial large flow-rate axial fan operated at 1770rpm was studied experimentally in laboratory conditions. The flow characteristics on the fan blade surfaces were investigated by measuring the pressure distributions on the blade suction and pressure surfaces and the results were discussed by comparing with analytical formulations and CFD. Flow visualizations were also performed to validate the flow characteristics near the blade surface and it was demonstrated that the flow characteristics near the fan blade surface were dominated by the centrifugal force of the fan rotation which resulted in strong three-dimensional flows. The time-dependent pressure measurement showed that the pressure oscillations on the fan blade were significantly dominated by vortex shedding from the fan blades. It was further demonstrated that the pressure distributions during the fan start-up were highly unsteady, and the main frequency variation of the static pressure was much smaller than the fan rotational frequency. The time-dependent pressure measurement when the fan operated at a constant speed showed that the magnitude of the blade pressure variation with time and the main variation frequency was much smaller than the fan rotational frequency. The pressure variations that were related to the vortex shedding were slightly smaller than the fan rotational frequency. The strain gages were used to measure the blade stress and the results were compared with FEA results.
The aim of this paper is to understand the time averaged pressure distributions and unsteady pressure patterns in a channel diffuser of a centrifugal compressor. Pressure distributions from the impeller exit to the channel diffuser exit are measured and discussed for various flow conditions. And unsteady pressure signals from six fast-response sensors in the channel diffuser are analyzed by decomposition method. The strong non-uniformity in the pressure distribution is obtained over the diffuser shroud wall caused by the impeller-diffuser interaction. As the flow rate increases, flow separation near the throat, due to large incidence angle, increases aerodynamic blockage and reduces the aerodynamic flow area downstream. Thus the minimum pressure location occurs downstream of the geometric throat, and it is named as the aerodynamic throat. And at choke condition, normal shock occurs downstream of this aerodynamic throat. The variation in the location of the aerodynamic throat is discussed. The pressure ratio waveforms by blade passing show regular oscillation not only for the normal but also for the surge conditions and the high frequency fluctuations are superposed on the oscillating pressure waveform as the flow rate increases. Periodic unsteadiness by blade passing does not decay in the diffuser channel. It depends on the operating point and is generally larger in the channel than in the vaneless space. Aperiodic unsteadiness rapidly decrease downstream of diffuser channel.
This paper describes a new topological clean up procedure to improve the quality of unstructured triangular meshes. As a post-processing step, the topological clean up procedure is applied both for elements that are interior to the mesh and for elements connected to the boundary and then Laplacian-like smoothing is used by default. Previous clean up algorithms are limited to eliminate the nodes of degree 3, 4, 8, 9, 10 and pairs of nodes of degree 5. In this study, a new clean up algorithm, which improves the quality of the triple connections with the degree of meshes 5 and 7 (i. e.; 5-7-5, 7-7-5, 7-5-7 etc.), is added. The suggested algorithm is applied to two example meshes to demonstrate the effectiveness of the approach on improving element quality in finite element meshes.
Developing a low NOX gas turbine combustor usually requires many experiments on flow and combustion. Computational prediction technology in the field of combustion has progressed with advances in computer technology and turbulent combustion models. This report covers applications of CFD for the development of the gas turbine combustor and some discrepancies between CFD and test results as well as improved methods for resolving such discrepancies.
The space-time conservation element and solution element (CE/SE) method, an accurate and efficient explicit numerical scheme for resolving moving discontinuities in fluid mechanics problems, is used to solve axisymmetric heat conduction problems with melting and/or freezing. The axisymmetric formulation is presented. Comparisons are made to existing analytical solutions. The CE/SE method is found to be accurate and robust for the numerical modeling of phase change problems.
In the present experimental study, the stability of Rayleigh-Bénard convection has been investigated in rectangular enclosures filled with microemulsion Phase-Change-Material (PCM) slurry. The PCM slurry exhibited a pseudoplastic non-Newtonian fluid behavior. Hysteresis in convection was clearly observed while the PCM were in a solid phase and in phase changing. The critical Rayleigh number decreases with the PCM mass concentration while the PCM is in phase changing. The fluid temperature at the center of the enclosure showed a time-dependent oscillation during the transition from a heat conduction state to a convection state. The maximum Nusselt number has been observed for all of the slurries while the heating plate was controlled at a temperature that most of PCM was in phase changing.
Thermal stability of a superconducting tape subjected to thermal disturbances under the effect of a two-dimensional hyperbolic heat conduction model is numerically investigated. The thermal inertia of the material, the finite speed of heat transfer, the temperature dependence of the Ohmic heat generation, and the finite duration and finite length of the thermal disturbances are taken into consideration in the mathematical model. Two types of superconductor tapes are considered, Types II and I. It is found that hyperbolic model predicts a wider stable region as compared to the predictions of its parabolic counterpart, the hyperbolic model shows a higher peak at early stages and the difference between the two models is significant at higher values of disturbance intensity and smaller values of time. The effects of different design, geometrical and operating conditions on superconducting tape thermal stability are also studied.