Hello,
I am using MFiX version 18.1.5 and I run my simulations on a cluster. I have implemented a simple Stokes drag with user-defined fluid velocity profile (that was calculated analytically). There is no fluid inlet/outlet – this fluid is simply assumed to be in the domain and imparts a drag on the particles. Particles (100 microns, 9 g/cc density, monodisperse) are initialized in a rectangular box with an initial solids volume fraction of 30% and they fall freely under gravity. The code seems to work just fine for this case. However, I want to test how this drag model works by monitoring a single particle freely falling under gravity and experiencing the drag force dictated by the particle-fluid relative velocity. This doesn’t seem to work and I see the following output for this case (right after I run). I have also tried debugging this with two particles randomly placed in the domain, but to no avail.
1 Like
First, you should update to the latest MFiX version (currently 19.2.2).
The negative number of DEM iterations looks suspicious. Not sure if it is related to the extremely small time step. Check the DT, TSTOP and particle properties.
For free falling particles, we usually use a cyclic BC in the y-direction (assuming gravity is acting in the negative y-direction, which is the default).