I am trying to reproduce Tukov
 

I am trying to reproduce Tukovic & Jasak work on Langrangian solver for the first Test case ( section 4.1 in thier article ), but using icoStructFoam, i placed the pillar 0.2 x 0.2 x 2 in a box 5.2 x 1.2 x 2.5 with base of pillar short end to (2.1,0.6,0). The grid was simple an cartisian grid, dx=0.1, the
Gambit solved this easily and imported by using fluentMeshToFoam. I mimic the procedure for icoStructFoam and alterad the required fields as described. The calculations works fine until moving the mesh, where the following error occur.

"Cannot determine normal vector from patch"

This at line 112 in twoPointCorrector.

I cannot understand the cause of the error!

Could someone please help me?

The problem was resolved. I fi
 

The problem was resolved. I figured it out by finding one of the 2D flag activated. 2Dmotion. Now it works nice.