Fluidization of a binary homogeneous mixture

Hello experts @jeff.dietiker @cgw . I am working on the fluidization of a homogeneous binary mixture of particles with different size(196 micron and 550 micron) and same density(2500 kg/m3) in a gas-solid fluidized bed. Minimum fluidization velocity of the mixture is around 5 to 6 cm/s but even at high inlet velocities (15 cm/s -20 cm/s),I am unable to see any movement of particles. I am using linear spring dashpot collision model. What should be the value of normal spring constant??

For kn=1000 N/m and normal coefficient of restitution 0.9, the collision time and DT_SOLID are close to 10^(-5) and 2*10^(-7) respectively. I am using a fluid solver time step of 10^(-5).

When I run the simulation, DEM solver time increases from 0.5-0.6 s to 7s-8s after few hours. Why is there no visible bubbling on the top surface or any movement at the core even at high velocities. I am attaching herewith the .mfx file. Could someone please guide me on this? Thanks in advance.
Homogeneousbed_10cms-1.mfx (17.0 KB)
particle_input.dat (631.1 KB)

I would be thankful if anyone could help me in this. Waiting for the response.

Do you really want to use the DEM model for this? By the size of it, it sounds more like TFM. You can add different particle sizes there when you have the same material/density. I believe the TFM model is one of the the most mature and stable models - and at least consiering the micron particle sizes you mention.

Actually, I want to visualise the movement of particles, which is possible with DEM. The actual setup contains 16 million particles. I have scaled it down to 7955 particles by keeping the particle diameters as same and simply reducing the setup dimensions.

Did you base your setup on some of the tutorials, with regards to IC and BC’s?

To create a uniformly mixed bed, I first created multiple bed regions to generate the particles. Then I disabled the fluid solver (pure granular flow) to let the particles settle and saved the particle positions at the end of the run. This output file was then used as the particle input for the simulation and boundary conditions were applied according to the three dimensional fluidized bed tutorial.