[Cucko]

Hello all,

I am a greenhorn at cfx so sorry for my stupid question. My Problem is that my geometrie is a rectangular microchannel with same cross section at Inlet and Outlet. I used the function AreaAve at Inlet and get 0.00318 m/s. At the Outlet i get 0.00331 m/s (after Grid study). My opinion is that this value is false, because of continuity equation it have to be the same value (density is the same) v1*A1=v2*A2 (A1=A2). If i use the ave Velocity function i get nearly the same value. So is the areaAve function the wrong for calculating velocitys at areas?

Sorry for my bad english and thanks for any help

Greetings cucko

[ghorrocks]

Is your simulation converged? It could be inadequately converged.

If your flow is incompressible with the same area for inlet and outlet then the areaAve velocity should be the same. What answer do you get when you use the massflow() function?

[Cucko]

Hi Glenn,

Thanks for ur quick answer.It was a simulation with water at 23°C (isothermal); so incompressible fluid. Its a laminar simulation (Reynoldsnumber at Inlet should be 2).
It is really freaky but i get with the function =massflow()@Inlet this value 1.000e-05 [kg s^-1] for the Outlet =massflow()@Outlet = -1.000e-05 [kg s^-1].
So this should be true. The Convergence was still ok and the Imbalance was nearly zero. I thought it could be a fault/artefact from the mesh. But my mesh is still ok i think. Min Angle >32° Determinant2x2x2 >0,7 Aspect ratio <10.
So i m really confused .

[ghorrocks]

Can you post an image of what you are modelling?

[Cucko]

Hi Glenn,

Sure. This is an Pic of a XZ Plane in the middle of the channel.

[ghorrocks]

I am not certain of this but I suspect the areaAve() function averages the nodal values over the control volume face, but massflow() uses the integration points to integrate the flow. This means massflow() is more accurate.