Hi,
I'm running a fluidized bed using cfdemsolverIB because my particles are too big (6 mm) for the mesh I need to solve the fluid phase. I have around 30 thousand particles into my system. When I try to run the case with few particles, the simulation doesn't present any problem. However, when I put all the 30 thousand particles into the system, it gives a Fatal Error after few timesteps. The error is:
--> FOAM FATAL ERROR:
Problem: edge would get 1 face using it only edge:478751 nFaces:1 vertices: (103857 158587) coords:(0.150004 0.141528 0.207) (0.154841 0.136401 0.1935).
It seems to be a problem in the solverIB mesh refinement. I'm wondering if this solver can be used for a large number of particles as I need. Is there any way to avoid this problem?
Thank you!
Gabriela Lopes.
alice | Mon, 09/18/2017 - 07:52
Hello Gabriela,
Hello Gabriela,
generally speaking, 30000 particles are a lot for resolved CFD-DEM. Considering that around 8 cells per particle diameter are required for a good representation within the mesh this would lead to a minimum of 15 million cells, if one assumes that the domain is not much bigger than the particle area. Is it possible that this quality criterion has not been met?
If you really want to resolve the particles precisely, please have a look at the following post: https://www.cfdem.com/forums/mesh-refinement (please scroll down to get to the important part.)
A small bug made its way into the last release and needs to be fixed (will be done in our next release, but int he meanwhile I'd reocmmend to do that yourself in your version. The fix is very simple...).
Cheers,
Alice
gacleq | Mon, 10/09/2017 - 06:26
Hi Alice,
Hi Alice,
Yes, you are right, I don't think it will be possible to use the solverIB in my case. The mesh would be extremely fine if I consider 8 cells per particle.
Thank you for your kind reply! Best regards,
Gabriela.