[gian93]

Hi Guys,

I'm simulating a simple steady state channel flow with a butterfly valve . I want to control the evolution of the computed mass flow rate at the outlet patch to evaluate the solution convergence (i also use the residual plot).


I used rhoSimpleFoam as solver.



Theese are the BC:


INLET

U >>> flowRateInletVelocity (I impose a mass flow rate of 0.5 kg/s)

p >>> zeroGradient



OUTLET



U >>> inletOutlet (in my case is the same as pressureNormalInletOutletVelocity)
P >>> fixedvalue (I impose a downstream pressure of 350000 Pa)





to calculate the mass flow rate at outlet patch I' ve added to my controlDict the functionObject for mass flow calculation :




faceObj1
{
type faceSource;
functionObjectLibs ("libfieldFunctionObjects.so");
enabled true;
outputControl timeStep;
log true;
valueOutput false;

sourceName Outlet;

operation sum;

fields
(
phi
);
}










THE QUESTION:



With this function object I would graph the mass flow rate at the outlet patch over each step iteration. I think that after few iteration the solver will find the solution and the mass flow rate at outlet will be equal (for mass conservation) to the inlet one (asymptotic trend). For the early time steps i would expect an oscillatory value of the variable phi (in this case mass Flow rate) that tends to stabilize when solution reach the convergence.

The problem is that this function object gives me a value of the outlet mass flow rate always equal (exactly equal) to the same value of the inlet patch without the expected initial oscillations...

Why i can't see the oscillations of the solver initial time step ?? There is something that i've missed from a phisical point of view in my BC or in the functionObject itself?
As it concerns inlet patch if i follow the Pressure evolution by an appropriate functionObject i can appreciate the "convergence trend".. but why it do not happen for mass flow rate at outlet?


thanks!!