Turbulent Mixing of Fuel and Air in Wake of Bluff Body Burner
I am trying to analyze the turbulent mixing of two gases: (1) air and (2) a fuel gas mixture with water vapor, methanol vapor, and a few other gas species. The flow region for the mixing that I am trying to analyze is the wake of a bluff body burner. (Please note that I am not trying to do the analysis for a situation where combustion is occurring.)
I read a forum post about the mixing of two gases in a pipe Turbulent mixing of two different gases, how? that was very helpful. I tried to emulate the simulation settings there as best I could, but I did not understand how the parameters for the Inflate boundary layer mesh refinement were set; so I went with a region refinement.
A graphical display of the mixing fraction in the wake of the bluff burner like that which which was given in the post above would be a great result for me. If possible a plot of average flow velocity in the wake of the bluff body burner would be fantastic.
I started the run of the simulation, but after 1.5 hours the mesh was still not generated. I would appreciate some help in implementing a better mesh refinement, and any other suggestions for analyzing the mixing of the two gases. Would passive scalar transport help?
Hello @lida1743 , thanks for posting your question in the Forum.
The minimum cell sizes and total number of cells are important meshing parameters when starting a CFD simulations. It can be difficult to decide your mesh configuration at design iteration 1 usually.
Hence, I always recommend performing a mesh sensitivity study just like presented in this documentation page. This method will allow you to perform a trade off between the computational cost and accuracy of the simulation. You will be preventing any unwanted computational usage like in your case as well. Following this documentation, please start with a mesh that is as coarse as possible without refinements, then start adding slight refinements one by one.
Thanks for the suggestion Kaany! I will try the simulation again this time with a mesh that is as coarse as possible without refinements. I have also reduced the size of the flow volume preceding and following the Bluff Body to (hopefully) reduce the number of cells.
I reduced the size of the flow volume and reran the simulation with the simulation using the standard mesh generated automatically. I checked the mesh quality metrics and they seemed to be acceptable; so I ran the simulation again. The simulation diverged, because the modified pressure was unrealistically large and was not decreasing.
A message I got at the end of the simulation suggested that I examine the meshing close to the point that was causing the trouble. The coordinates of the point were close to x = 0.07m y = -0.02m z = 0.9m. I am unsure of how best to refine the mesh, and I am hoping you can offer a suggestion.
Thanks,
David
