[mturcios777]

Hello everyone,

I've been doing some experimenting with dynamic refinement in compressible solvers and have found a problem I am not sure how to deal with. I am using the dynamicRefineFvMesh on the aachenBomb which has been modified with a coarser starting geometry and a sprayDyMFoam solver which is a synthesis of rhoPimpleDyMFoam and sprayFoam.

For some reason, after solving rhoEqn, the density in the refined cells increases dramatically to the extent that the the solver eventually crashes. I have attached an image of the rho, p, U and C7H16 fields immediately after rhoEqn.H. I confirmed that rhoEqn.H is the cause as I also wrote out the fields just after mapping onto the refined mesh and before rhoEqn and the fields don't suffer the discontinuity.

Anyone have any ideas as to what could be causing this? It seems like the ddt(rho) term in the equation is what is causing the problem. I've included the dynamicMeshDict, fvSchemes and fvSolutions files for reference.

Thanks,