I think I know what's going wrong but I am surprised that you're running into this issues! It seems like we have a problem in the residual projection if the Helmholtz iterations are ZERO. Please disable the residual projection for now (set p94 and p95 to zero) and check what happens if you integrate for say 1 flow-through time. Just for completeness: Can you post your .usr file as well. Stefan On Jan 19, 2010, at 10:35 AM, [email protected] wrote:

I created the mesh with PRENEK. Yes, in x I hace periodic bcs, wall in y (top and bottom) and I also used periodic bcs in z. For USERBC I used the same like in the turbChannel example.

_______________________________________________ Nek5000-users mailing list [email protected] https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users