The eruptive regime of mass-transfer-driven Rayleigh–Marangoni convection

2016 ◽  
Vol 791 ◽  
Author(s):  
Thomas Köllner ◽  
Karin Schwarzenberger ◽  
Kerstin Eckert ◽  
Thomas Boeck

The transfer of an alcohol, 2-propanol, from an aqueous to an organic phase causes convection due to density differences (Rayleigh convection) and interfacial tension gradients (Marangoni convection). The coupling of the two types of convection leads to short-lived flow structures called eruptions, which were reported in several previous experimental studies. To unravel the mechanism underlying these patterns, three-dimensional direct numerical simulations and corresponding validation experiments were carried out and compared with each other. In the simulations, the Navier–Stokes–Boussinesq equations were solved with a plane interface that couples the two layers including solutal Marangoni effects. Our simulations show excellent agreement with the experimentally observed patterns. On this basis, the origin of the eruptions is explained by a two-step process in which Rayleigh convection continuously produces a concentration distribution that triggers an opposing Marangoni flow.

1997 ◽  
Vol 119 (4) ◽  
pp. 711-719 ◽  
Author(s):  
M. J. Braun ◽  
M. B. Dzodzo

The laminar flow in a hydrostatic pocket is described by a mathematical model that uses the three-dimensional Navier-Stokes equations written in terms of the primary variables, u, v, w, and p. Using a conservative formulation, a finite volume multiblock method is applied through a collocated, body fitted grid. The flow is simulated in a shallow pocket with a depth/length ratio of 0.02. The flow structures obtained and described by the authors in their previous two dimensional models are made visible in their three dimensional aspect for both the Couette, and the jet dominated flows. It has been found that both flow regimes formed central and secondary vortical cells with three dimensional corkscrew-like structures that lead the fluid on an outward bound path in the axial direction of the pocket. In the Couette dominated flow the position of the central vortical cell center is at the exit region of the capillary restrictor feedline, while in the jet dominated flow a flattened central vortical cell is formed in the downstream part of the pocket. It has also been determined that a fluid turn around zone occupies all the upstream space between the floor of the pocket and the runner, thus preventing any flow exit through the upstream exit of the pocket. The corresponding pressure distribution under the shaft for both flow regimes is presented as well. It was clearly established that both for the Couette, and the jet dominated cases the pressure varies significantly in the pocket in the circumferential direction, while its variation is less pronounced axially.


2014 ◽  
Vol 2014 ◽  
pp. 1-4 ◽  
Author(s):  
Zujin Zhang

We consider the three-dimensional Boussinesq equations and obtain a regularity criterion involving the pressure gradient in the Morrey-Companato spaceMp,q. This extends and improves the result of Gala (Gala 2013) for the Navier-Stokes equations.


1999 ◽  
Vol 121 (2) ◽  
pp. 460-468 ◽  
Author(s):  
A. Mukhopadhyay ◽  
P. Venugopal ◽  
S. P. Vanka

A three-dimensional numerical simulation of linearly sheared flow past a circular cylinder has been performed for a shear parameter β of 0.02 and a mean Reynolds number of 131.5. A cylinder of 24 diameters span is considered. A second-order accurate finite volume scheme is used to integrate the unsteady Navier-Stokes equations. Present computations confirm both qualitatively and quantitatively, the aspects of cellular shedding as reported by several investigators through experimental studies. Up to five constant frequency cells of obliquely shedding vortices are observed. The nondimensional frequencies of these cells are observed to be lower than those given by parallel shedding correlations at the equivalent Reynolds numbers. It is also observed that the cell boundaries continuously move in time. Detailed distributions of vorticity and velocity components are presented to describe the flow. The influence of end-wall boundary conditions is studied by computing two cases, one with free-slip condition, and the other with no-slip condition on disks of radius of five cylinder diameters.


Fluids ◽  
2019 ◽  
Vol 4 (1) ◽  
pp. 40 ◽  
Author(s):  
Dajun Liu ◽  
Takafumi Nishino

A series of three-dimensional unsteady Reynolds-averaged Navier–Stokes (RANS) simulations are conducted to investigate the formation of stall cells over a pitching NACA 0012 aerofoil. Periodic boundary conditions are applied to the spanwise ends of the computational domain. Several different pitching ranges and frequencies are adopted. The influence of the pitching range and frequency on the lift coefficient (CL) hysteresis loop and the development of leading-edge vortex (LEV) agrees with earlier studies in the literature. Depending on pitching range and frequency, the flow structures on the suction side of the aerofoil can be categorized into three types: (i) strong oscillatory stall cells resembling what are often observed on a static aerofoil; (ii) weak stall cells which are smaller in size and less oscillatory; and (iii) no stall cells at all (i.e., flow remains two-dimensional) or only very weak oval-shaped structures that have little impact on CL. A clear difference in CL during the flow reattachment stage is observed between the cases with strong stall cells and with weak stall cells. For the cases with strong stall cells, arch-shaped flow structures are observed above the aerofoil. They resemble the Π-shaped vortices often observed over a pitching finite aspect ratio wing.


2008 ◽  
Vol 130 (3) ◽  
Author(s):  
F. Abbasian ◽  
J. Cao ◽  
S. D. Yu

Three-dimensional oscillatory airflows induced by a rotating grass-cutting blade in a cylindrical chamber are studied experimentally and numerically in this paper. Experimental pressure results are obtained using a sound pressure transducer and a data acquisition system. The measured pressure data contain background noise and high-frequency sound signals due to the blade vibrations. The background noise is separately measured; its effect on the signal is determined from a spectral subtraction algorithm. A time-accurate finite volume numerical solution to the three-dimensional incompressible unsteady Navier–Stokes equations is also sought using the sliding frame technique and the unstructured tetrahedral mesh. Convergence studies are conducted using various combinations of mesh sizes and time increments to ensure the stability of the numerical scheme. The experimental and numerical pressure results are in good agreement.


1994 ◽  
Vol 116 (1) ◽  
pp. 116-123 ◽  
Author(s):  
G. Freskos ◽  
O. Penanhoat

The demand for efficiency in today’s and in future civil aircraft is such that experimental studies alone do not suffice to optimize aircraft aerodynamics. In this context, much effort has been spent in the past decade to develop numerical methods capable of reproducing the phenomena that occur in the engine flow field. This paper presents some studies in Computational Fluid Dynamics related to supersonic inlets. Two approaches are considered. First, there is a need for a code capable of calculating in a cost-efficient way the entire flow field around a two-dimensional or three-dimensional inlet, e.g., to perform parametric studies. To this effect, a computing method based on grid construction by mesh generator dedicated to inlet shapes and on the discretization of the unsteady Euler equations with an explicit upwind scheme was developed. The treatment of complex geometries led us to adopt a multiblock grid approach. Therefore particular attention was paid to the treatment of the boundary conditions between the different domains. Second, there is a need for a code that can capture local phenomena in order to get a better understanding of inlet behavior (shock/shock, shock/boundary layer interactions, etc.). To this effect a two-dimensional turbulent Navier-Stokes code is used. The two-equation k-ε turbulence model included in the program seems to be one of the most successful models for calculating flow realistically. Correction of the near-wall influence extends its capability to complex flow configurations, e.g., those with separated zones.


1997 ◽  
Vol 119 (3) ◽  
pp. 569-576 ◽  
Author(s):  
L. Bai ◽  
M. Fiebig ◽  
N. K. Mitra

Numerical simulation of three-dimensional unsteady turbulent flows in fluid couplings was carried out by numerically solving Navier-Stokes equations in a rotating coordinate system. The standard k-ε model was used to take turbulence into account. A finite volume scheme with colocated body-fitted grids was used to solve the basic equations. Computed flow structures show the vortex generation and its effect on the torque transmission. Computed local velocity and torque flow compare well with measurements.


2014 ◽  
Vol 743 ◽  
pp. 75-123 ◽  
Author(s):  
Georg F. Dietze ◽  
W. Rohlfs ◽  
K. Nährich ◽  
R. Kneer ◽  
B. Scheid

AbstractFull numerical simulations of the Navier–Stokes equations for four cases of vertically falling liquid films with three-dimensional surface waves have been performed. Flow conditions are based on several previous experimental studies where the streamwise and spanwise wavelengths were imposed, which we exploit by simulating periodic wave segments. The considered flows are laminar but approach conditions at which intermittent wave-induced turbulence has been observed elsewhere. Working liquids range from water to silicone oil and cover a large interval of the Kapitza number ($\textit {Ka}=18\mbox{--}3923$), which relates capillary to viscous forces. Simulations were performed on a supercomputer, using a finite-volume code and the volume of fluid and continuum surface force methods to account for the multiphase nature of the flow. Our results show that surface waves, consisting of large horseshoe-shaped wave humps concentrating most of the liquid and preceded by capillary ripples on a thin residual film, segregate the flow field into two regions: an inertia-dominated one in the large humps, where the local Reynolds number is up to five times larger than its mean value, and a visco-capillary region, where capillary and/or viscous forces dominate. In the inertial region, an intricate structure of different-scale vortices arises, which is more complicated than film thickness variations there suggest. Conversely, the flow in the visco-capillary region of large-$\textit {Ka} $ fluids is entirely governed by the local free-surface curvature through the action of capillary forces, which impose the pressure distribution in the liquid film. This results in flow separation zones underneath the capillary troughs and a spanwise cellular flow pattern in the region of capillary wave interference. In some cases, capillary waves bridge the large horseshoe humps in the spanwise direction, coupling the two aforementioned regions and leading the flow to oscillate between three- and two-dimensional wave patterns. This persists over long times, as we show by simulations with the low-dimensional model of Scheid et al. (J. Fluid Mech., vol. 562, 2006, pp. 183–222) after satisfactory comparison with our direct simulations at short times. The governing mechanism is connected to the bridging capillary waves, which drain liquid from the horseshoe humps, decreasing their amplitude and wave speed and causing them to retract in the streamwise direction. Overall, it is observed that spanwise flow structures (not accounted for in two-dimensional investigations) are particularly complex due to the absence of gravity in this direction.


2021 ◽  
pp. 440-445
Author(s):  
Kyrylo S. Krasnikov

To modify easily chemical consistence of molten steel metallurgical plants usually add alloys in a form of lumps through a hopper at ladle`s top. The problem is a large variety of adjusted technological conditions for this process, which leads to iterative discovery of rational ones. The article presents synthesis of a mathematical model for the mentioned process using Navier-Stokes and Euler-Lagrange equations. It is designed for researching of process modes depending on: a horizontal position of addition hopper relatively to argon plug at ladle bottom, an average diameter of spherical addition lumps, a total mass of addition heap, a necessity of the second hopper, a temperature of the melt. It takes into account interconnected three-dimensional fluid and solids dynamics, temperature exchange between melt and lump, also it computes level of concentration homogenization. A numerical experiment shows a significantly better addition homogenization when two hoppers are simultaneously used for a feeding. Adequacy checks are performed using ice balls in air-stirred water.


Sign in / Sign up

Export Citation Format

Share Document