[hasanmas]

Hallo every one,

i am really hoping and looking forward for some help!

basically i am having a hard time making my simulation of a multi-phase with a porous medium converge or even work.

so far this is what i have tried:

the simplified model is basic a mix between angledDuctExplicit (geometry/porous medium) and vertical (Lagrangian/model and solver/ sizeDistribution: RosinRammler - no chemistry) channel tutorial. three dummy points were picked to be used a injectors (coneInjection).

using the LTSreactingParcelFoam on a simplified model:

1. the simulation converges on a rough mesh but when using a finer mesh no converges!

2. on the finer mesh the simulation does not converge even after changed or switching on/off all of the fallowing:

• porous medium.
• number of parcels
• turbulence mode (on/off/laminar/sst-k-omega)
• sub models of the reaction model (gravity/drag).

3. initialization of the simulation using a mapfield + another solver didnt help!


4. using a different thermo or transport model didnt help!





using the reactingParcelFoam on a simplified model:


1. the simulation crashes as soon as one parcel injected.


2. initialization of the simulation using a mapfield + another solver didn't help!


3. using a very small/smaller time step didn't help!



using either LTSreactingParcelFoam/reactingParcelFoam on the real model:

using all of the the variations:

the model either cashes or when i am really really really really really really lucky: the simulations stops and gets stuck after adding new parcels.
-------------------------------------------------------------------------------


given the bad documentation of the reactingCloud1properties file i think it would be helpful if any one can explain a way to understand how many parcels are introduced and how to calculate that number form the input parameters(massFlowRate/massTotal/duration/parcelsPerInjector/Umag/parcelsPerSecond).

does any one have any ideas?