Looking at the reference for the Di Felice drag correlation ( Di Felice, R. The Voidage Function for Fluid-Particle Interaction Systems. International journal of multiphase flow,1994, 20 (1), 153–159.): the Dalla Valle (1948) correlation was used for drag force on isolated spheres. However, the Schiller & Naumann (1933) correlation is used in the MFiX source code.
Recall that the Di Felice drag law correlates the Ergun equation for closely packed solids togehter with experimental data for lower solids fractions, to apply to the full range of solids fractions. Using Dalla Valle correlation in place of Schiller & Naumann would appear to get the Di Felice correlation closer to the Ergun & Wen-Yu correlations at higher and lower solids fractions respectively. Please see the comparison table below (for 63 micron particles in ambient air flows).
Are you suggesting MFiX should be changed to use the Dalla Valle correlation? Or a new option (keyword) that would control which correlation equation is used? Are you able to contribute code to implement this?
Considering that the De Felice drag law was correlated using Dalla Valle, I believe that MFiX should use the same too. However, I have only checked it for one scenario (Geldart A particles at particle Reynolds numbers < 100) and am not sure how it compares at other scenarios. There is one publication ( Wang, T.; Wang, S.; Shen, Y. Particle-Scale Study of Gas-Solid Flows in a Bubbling Fluidised Bed: Effect of Drag Force and Collision Models. Powder technology, 2021, 384, 353–367) where this correlation (with Dalla Valle) was found to give good results for bubbling beds.
In case Dalla Valle is to be used in place of Schiller & Naumann, then the change in code is trivial I guess. I have updated the attached ‘drag_gs.f’ file (lines 2162 to 2170) for the same. Appreciate if you would confirm that it is in order and that this is the only modification required. drag_gs.f (81.5 KB)
