Monitor engineering data - Mflux not consistent?
 

Hi,

I have a simple model of a duct with variable cross-section area. Along its length, I've defined several shells which represent cross sections.
If I load the converged solution and calculate the mass flow rates through Post >> Fluxsum >> Shell type..., I get the same value for all shells, and that's how it should be.
However, if I define an internal boundary at these cross sections and let Mflux be monitored, the values I get in the *.erd file make no sense at all (not in kg/s nor in kg/( m2 s) ).
I'm pretty sure the internal boundaries are defined correctly - all faces are oriented to the same direction and so on. The Mflux (in [kg/s]) are consistent at the inlet and outlet with the boundary conditions. But the ones within the solution domain are not.
Does anyone know what is going wrong?