[John Lee]

Dear all,

I have problem simulating a channel flow over some layers of particles using a CFD+DEM model.

My model is like this: On the bottom of a flume, I place a certain thickness of randomly-packed particles. Then I assign the inlet as a velocity boundary and the outlet as a pressure outlet.
The purpose is to: study the sediment erosion in the channel.

The problem is: Although the model seems running successfully and particles start to move when the fluid velocity reaches a certain magnitude, the simulation results are suspicious to me. At least, I can't validate them with existing experimental studies. In my simulation results, the fluid velocity within the particles never reaches zero. I do expect that there is some internal flow within the particles. However, the internal flow should be within some distance below the top surface, beyond which the fluid will be difficult to penetrate. But based on my simulation results, the fluid velocity field within the particle bed is some kind of chaos and never dissipates gradually to zero. It feels like that these solid particles have no blocking effect for fluid penetration.

Any idea on this problem is greatly appreciated. Thanks for your time!

Sincerely,
John

[me3840]

Are you running double-precision? If you're just getting noise in the bed that's what you may be seeing. What is the velocity magnitude in the bed?