[ajjadhav]

Hello,
I am simulating the solid-liquid flow in a horizontal pipe using the Eulerian-Eulerian approach.
Details are as follows:
Pipe length: 2m
Pipe diameter: 40 mm
Liquid phase: Sugar Solution (Average velocity: 0.72m/s, density: 1140kg/m3 and viscosity: 0.0044 kg/m-s)
Solid phase: Alginate beads (Average velocity: 0.74m/s, size: 2mm, density: 1150kg/m3 and volume fraction:0.063)

When I use the k-Epsilon model it predicts solid distribution very well but not velocity profile (underestimate at the center of pipe), whereas if I use the SST model it predicts velocity profile very well but not solid distribution (overestimate at center).

Do you know how can I resolve this issue?

Thank you

[ghorrocks]

The turbulence models provided with CFX are tuned for single phase flows. So there are no guarantees they will work well for multiphase flows - this is an active area of research and there are many multiphase flows where nobody has designed a good turbulence model for it.

So you will get results like this from time to time. My suggestions are:
* Make sure your simulation is converged and grid independent (by a mesh sensitivity check)
* Read the CFX documentation on multiphase modelling and especially turbulence modelling in multiphase. There are lots of options available and some of these options might be relevant to you case.
* Consider what is the physics by which the particles are affecting the turbulence and the flow, and choose a turbulence model which best captures those physics.