I tested 3D grid refinement with immersed boundary and two different dynamics recently. The two dynamics defined as below:
Dynamics<T,DESCRIPTOR> *dynamics = new ConsistentSmagorinskyCompleteRegularizedBGKdynamics<T,DESCRIPTOR>(param.omega, param.cSmago); Dynamics<T,DESCRIPTOR> *dynamics = new ConsistentSmagorinskyBGKdynamics<T,DESCRIPTOR>(param.omega, param.cSmago);
The results of the two are obviously different. The force with ‘CompleteRegularized’ one is twice as big as the other one. The only difference in my code is the definition of dynamics shown above.
I wonder why it’s exactly twice as much, rather than a similar result.