[uosilos]

Hi,

I have tried to compile a lagrangian solver based on coalChemistryFoam adding MPPIC and Colliding features in OF8.

When I run the solver it only recognizes as dispersionModel "none".

I noticed we have /intermediate/submodels/kinematic/dispersionmodel where the noDispersion is places, and the turbulence/submodels/kinematic/dispersionmodel where we have the DispersionRASModels: StochasticRASDispersion and GradientDispersionRAS.

In the own definition of DispersionRASModel it includes DispersionModel file,
Does anybody know which is the difference between the two "Dispersion" styles.


Thanks

[uosilos]

Originally Posted by uosilos

Hi,

I have tried to compile a lagrangian solver based on coalChemistryFoam adding MPPIC and Colliding features in OF8.

When I run the solver it only recognizes as dispersionModel "none".

I noticed we have /intermediate/submodels/kinematic/dispersionmodel where the noDispersion is places, and the turbulence/submodels/kinematic/dispersionmodel where we have the DispersionRASModels: StochasticRASDispersion and GradientDispersionRAS.

In the own definition of DispersionRASModel it includes DispersionModel file,
Does anybody know which is the difference between the two "Dispersion" styles.


Thanks

[oswald]

Roughly speaking, both do the same regarding the magnitude of the perturbance. The direction is either opposite to the direction of grad(k) for gradientDispersion and in a random direction for stochastic dispersion. I think this is also described in the headers of the "*DispersionRAS.H"-files. Hope this helps!

[uosilos]

Originally Posted by oswald

Roughly speaking, both do the same regarding the magnitude of the perturbance. The direction is either opposite to the direction of grad(k) for gradientDispersion and in a random direction for stochastic dispersion. I think this is also described in the headers of the "*DispersionRAS.H"-files. Hope this helps!

Thanks a lot Oswald.


I refer to the fact that they are kind of split in those two directories (lagrangian/intermediate and lagrangian/turbulence), with "creating" files being quite different (dispersionModel vs dispersionRASModel), more than to their theoretical nature.



Unfortunately this structure is not ideal for developing new classes...