[cissystef]

Hi everyone,

I am building up a simple case trying to simulate the pressure drop across a porous plate, and to see if it matches with the experiment.

Geometry:

The geometry is simple, a duct (0.6*0.6*3) and a plate (0.6*0.6) inside of the duct.

0 folder - Initial conditions:

the most important 3 parameters for setup:

U: walls (fixedValue; uniform (0 0 0));
inlet (fixedValue; uniform (0.88 0 0));
outlet (zeroGradient);
porous_half0/1 (cyclic) <-- generated by createBaffle

T: walls and outlet (zeroGradient)
inlet (fixedValue, uniform 300);
porous_half0/1 (cyclic) < -- generated by createBaffle

p_rgh: walls and inlet (fixedFluxPressure, uniform 0)
outlet (fixedValue, uniform 0)
porous_half0/1 (porousBafflePressure, cyclic, uniform 0 - jump, D = 66300, I = 100, length = 0.1) <-- generated by createBaffle


So the steps I took to run the case is: blockMesh --> createBaffle -overwrite --> buoyantBoussinesqSimpleFoam


I could run the case, but the residuals are not good, most of them are in between 0.01 - 1.

I have played with a lot of boundary conditions, still haven't solved the convergence problem. I can reduce the inertial coefficient, which seem to converge properly except the temperature, but the results are not matching with the experiment.

The inertial coefficient I calculated is based on a predicted line obtained from the experiment, the porous baffle pressure jump Delta P = - (D*mu*U + 0.5*I*rho*U^2)*L, so using a predicted line can obtain the coefficient of U and U^2, which gives the value for Darcy's coefficient and inertial coefficient.

Can anyone give me some insights into this problem?