OCTOBER 19-20, 2012 - YMCA University of Science & Technology

Proceedings of the National Conference on
Trends and Advances in Mechanical Engineering,
YMCA University of Science & Technology, Faridabad, Haryana, Oct 19-20, 2012
• Effects of gravity was accounted for to act in the negative z-direction
• Multi phase simulations were carried out.
• The pipe wall roughness was taken smooth
• A non-slip boundary condition was imposed at the stationary walls, so that velocity at the wall is forced to be
zero.
The different model parameters have been defined as close as possible to the actual experimental
conditions and Table 1 shows the selection of model parameters.
Parameter selection
Table 1 Selection of different simulation parameters
Solver
Formulation
Time
Space
Velocity formulation
Gradient option
Multiphase
Turbulent model
Near-wall treatment
k-ε multiphase model
Segregated
Implicit
Un Steady
3D
Absolute
Cell-based
Eulerian/ Mixture
Standard k-ε (2-equation)
Standard wall function
Per phase in Eulerian
In the above simulation specifications, The Eulerian and mixture model is used in Fluent software with standard
equations. The discretization method called ‘Phase Coupled SIMPLE’ was selected for the pressure-velocity
coupling while ‘First Order Upwind’ discretization method was used for other scalar parameters like momentum,
volume fraction, turbulent kinetic energy etc. The simulations have been carried out for different test conditions
in terms of solid loading ratio, superficial air flow velocity, and volumetric solid concentration and pressure drop
values. Inlet boundary conditions such as inlet velocity, volume fraction of solid etc, have been defined
according to the experimental parameters. Distribution of cross sectional velocity is reasonable to assume
uniform at the bend inlet. At wall, no slip condition is assumed and the wall roughness constant was taken as 0.5.
For the convenience of the simulations, spherical mono sized particles were assumed for all solid materials and
with this assumption; particle mean diameters were used for the simulations. To define the boundary conditions
at the inlet, the velocities of all the phases have to be given. The process of solving a multiphase system is
inherently difficult, and usually one may encounter some stability or convergence problems. After each
simulation, the velocity and pressure profiles and distribution of solid volume fractions of each phase were
inspected. The variations of the above variables with the time were also examined according to the simulation
results. Finally, pressure drops across the considered sections for different test conditions were calculated using
simulation results and then compared with the experimental observations.
Grid adaptation for y+
It is essential to make sure that the depth of the wall-adjacent cells fall within the distance over which the loglaw
is valid (30