CFD Online Discussion Forums - Turbulent Dispersion for Coal Combustion

Hi all,

I am looking at the possibility to use the model for Turbulent Dispersion for Lagrangian Tracking in my Ph.D. project which involves coal combustion.

I have read a number of articles on PF combustion using Fluent, where the turbulent dispersion model due to Gosman and Ioannides is included. As I understand this model is used in both CFX and Fluent, however I do not know if there are any differences in the implementation between the codes.

I have been in contact with a Ph.D. student who has done coal combustion simulations in Fluent using this turbulent dispersion model, and he did not recall any specific convergence issues related to using this model. From reading the CFX 12.1 documentation, it seems to me that applying the turbulent dispersion model is difficult task in regard to convergence behavior.

I have tested including turbulent dispersion to the Coal combustion tutorial included in CFX 12.1, but I did not manage to obtain a converged solution. However, I did not try to vary any parameters except for the Eddy Viscosity Ratio Limit which I increased step-wise up to 50 (default=5).

I have also tried simulating the turbulent dispersion for my own project, just using non-reacting particles with a reduced mass flow, which converges fine with just mean dispersion. I did not manage to obtain a converged solution for this case either.

My question is directed to anyone who has experience on this topic, if you have any recommendations what approach to take when including the turbulent dispersion model? Any model parameters that could be changed in a certain manner? Increasing the number of particle tracks, compared to what is sufficient for a mean particle dispersion simulation?

I would highly appreciate any thoughts on this.

Best regards,
Reine