I have created this model with particles (Diameter of 100 micron).It is running. Please have a look if the setting looks ok. This is a continuation from the topic “Unable to run MFiX case : Single body single-phase flow” that was resolved by Dr. Dietiker. I implemented the DEM related model in that. Please let me know if everything looks ok in this model and the solid boundary conditions are correctly done. The particles are supposed to come up along the tube. But they seems to be disappearing. How to make sure that the particles move up from the cup to the top along the tube. I have attached the Zip files and few figures and a movie file.
When you define the vtk region for the particle data, use the “Background” region so the vtp file will have particles in the entire domain. You used the “Particle initial position” region, which only saves particles in that smaller region. Delete the current particle vtk setup and add a particle data vtk region that uses the Background region.
Thank you so much for your help and suggestions. It worked perfectly and now I can see the particles everywhere. The figrue is attached. I will keep you posted going forward.
Is there any way possible to increase the resolution of the particles in the video and figures. or it will be original size always. Please let me know.
In your situation, I would rather open the results in Paraview. It allows you to scale up the representation of the particles, so you can easily visualize them.
I have uploaded the MFiX file and all other files to google drive and shared the link
I also tried to plot the .vtp files in Paraview. However, the particles are very different here. I have uploaded the screenshot as well as the particle movie file with MFiX. The particle plots in Paraview is very different from what I can see in MFiX (see attached movie file).
Kindly let me know which files I need to plot in Paraview to see the unsteady particle trajectories.
I think there is a tutorial on MFiX documentation for this.
In any case, in Paraview, open your .vtp main file (which is normally linked to all of them). Then create a glyph (see below). Glyph type, sphere. Scale array - Diameter. The value in the Scale Factor (see also below) should allow you to scale up and down your particles. Then, if you want to see the walls you may have to also import the .stl file. I let you give it a try.
I tried as per your instructions to use Paraview. I am attaching the screenshots of the Paraview and the list of files that I have incluing .vtp files. I also used the glyph and loaded the stl file as well. The particles show up inside the cup. Let me know if this is what I will be seeing. Another issue is this simulation is for 0.5 seconds only. When I click on the forward arrow to go to the next timestep, it goes to 1. I am trying to set the time increment to 0.1 but unable to do so.
Kindly advice how I can see the particle positions every 0.01 seconds.
I took a look at your mesh, and it needs to be improved. I have mentioned several times the stl geometry needs to extend past the MFiX box at inlets and outlets so you have a cleaner intersection. The mesh always need to be inspected before running the simulation. Since you did not extend the stl geometry past the MFiX box, you get a jagged intersection:
Thank you. I will certainly do that. Atleast I got a solution and it ran. I will modify everything and run again.
Now I have to focus on this Ellipta inhaler geometry.
I will open a new case and upload the MFiX file once I develop one for this Ellipta Inhaler Geometry. I am now in the process of creating the STL file.
I applied the truncation and generated a finer grid (160x320x160) and that smoothened out the mesh significantly. I am now starting a simulation with 50 microns particles. Let’s see how it goes. I will keep you updated.
I have modified the grids. I have created 3 levels of grids.
Medium : 80x160x80 = 1,024,000 mesh
Fine : 120x240x120 = 3,456,000 mesh
Finer : 150x300x150 = 6,750,000 mesh
Seems like the mesh is much more smooth now and the holes have been taken care of.
I am using the medium mesh for 150 micron carried particle, the fine mesh for 100 micron carrier particle, and the finer mesh for 50 micron carrier particle.
I will keep you posted how the simulations are going.
I was able to run the simulation for 80x160x80 (1,024,000) mesh with the inlets and outlets truncated. The simulation ran for 0.5 seconds. The files are located in google drive. Here is the shared link. I have also uploaded the movie file and the variation in the particle numbers at 3 locations. One thing I saw in the movie file is that the particles are not reaching the top. The top is pressure outflow. Do I need to specify any boundary conditions at the top for the particles as well (like freeze). I did not see any boundary conditions for particles at the outflow boundary. The particle number variation is attached. At t=0, there were 1016 particles.
Please let me know if a outflow boundary condition for particles needs to be specified.
I can see particles moving to the top in your movie. The total number of particles decrease over time, which indicate they leave the system. Unless they go through the wall (if the spring stiffness is too small), they must be exiting at the top. The monitor you use at the top to count the particles is not appropriate because it is a plane and it is unlikely you will get any particle center exactly at that plane location at any time. A better measure would be a mass flow rate monitor, but it could also miss particles if the flow is dilute.
Thank you. These are the csv files for the particle number trackings. I understand. At least the background and initial positions file show the variation of the particle numbers. You can close this case as solved. I will open a new case for the Ellipta Inhaler. It is a more complicated case but I am trying to set up the model. I will keep you posted.
I want to get the initial position coordinates of the particles at t=0 as explained in 4.6.11 in the website 4.6. Solids — MFiX 25.3 documentation. How can I export the initial position coordinates as a .csv or .dat file. Please let me know. I am also looking at a 2 solids case with Carrier and PI particles.
Please let me know how I can generate the initial position coordinates for the particles once it has seeded in the box.
