[greg93]

Hello dear all,

I'm working with multiphase simulations in hydrocyclones using Fluent for my Ph.D. thesis. I have some experience with CFD simulations but I'm still learning about multiphase simulations for solid-liquid interaction. For dilute suspensions, I'm trying to use the one-way coupling, and as the equipment is a hydrocyclone I use RSM to model turbulence. First I start the transient solution with water only. When some responses start to stabilize, such as Pressure Drop and outlet mass flow rates, I enable the DPM in the postprocessing step to release particles from injections. I know that the DPM is very sensitive to parameters such as Stochastic Tracking (in turbulent flows) and I tried to use some standard values as there are no detailed procedures for hydrocyclones in literature. However, no matter what I do I always get poor results for particle collection. For example, In my case particles of 5um should be collected with 50% efficiency, and when I track these particles in the solver I get 96%. For smaller particles, this efficiency reduces but is still around 90%.

Does anyone here has had the same problem when working with DPM? I know that this is a well-known approach but I still can't solve this issue.

I'm using a mesh with 226K elements which is a good size according to the GCI that I calculated. My hydrocyclone is small, with Dc=30mm.

I would really appreciate if you could help with this problem guys. I've been trying to search and study with the Ansys Users Guide and some documents that I found on google but I didn't succeed, and this is an important step in my Ph.D. research.