Hey! I have a 2D problem (simulated by a thin 3D problem) were i have got the following error in CFX-solver! But since i have no "free slip" bounderies, can anyone tell me what the problem could be?? The mesh is generated in Icem!

Jesper

"ERROR #002100048 has occurred in subroutine SU_BNEXT. Message: All vertices for a fluid domain lie on a boundary. The solver considers this to be a fatal error because control volume gradients cannot be calculated, leading to serious discretization error. | | This error may arise for two-dimensional simulations when a free slip boundary condition is applied to the end planes. In this situation a symmetry condition should be applied to the end planes instead. Execution is terminating. This error message can be bypassed by setting the expert parameter 'boundary vertex check = f'."

Did you set both end planes as symmetry condition?

No i did not! I have 3 inlet (top, buttom and front) and an outled at the end, and the sides are set as walls as same as the body! Why should that create the problem?? But you are right - If i set the sides as symmetri the solver has no errors!

Jesper

Execution is terminating. This error message can be bypassed by setting the expert parameter 'boundary vertex check = f'."

this sounds funny to me, if the execution is terminated in next step, what is the use of bypassing the warning. Anyway the solver will come out.

If you put wall boundary on the side planes, this is not a 2-D approximation.

You can bypass the warning if you set the problem up this way intentionally. Clearly the he did not.

Hi Jesper,

If you specify the side walls as "No Slip", there is nowhere for the fluid to go. A two dimensional problem is either symmetric or periodic in the 3rd dimension and requires boundary conditions accordingly.

Regards, Robin

I think the warning means that if you bypass than the solver will not exit, and will try to continue. Then it makes sense, what do you think.

This can also happen if all the vertices are on boundary conditions like inlets/outlets. I would think this is implied by the first sentence: "All vertices for a fluid domain lie on a boundary".

The example is just a common setup error.

Bypassing the warning only makes sense if you had intentionally set the problem up in this manner. Based on what has been stated, the user forgot to specify symmetry planes or periodic interfaces on the side walls. If he fixes this, there will be no warning and no need to bypass it.

You can just use local parallel run to cover this problem, I don't know why this will run properly without any warning message by local parallel run.