[jywang]

Dear developers and all users,

In the paper "Optimal Shape Design for Open Rotor Blades - AIAA 2012-3018", the authors have done the RANS simulation for Caradonna&Tune case using 3d periodic boundary, the mesh topology is shown in the first figure.

In order to test the 3d periodic boundary, I have built a computational field as the topology described above. The mesh is shown in the second figure in attachment, and the procedures how I run the this simulation are:

1. Output the su2 mesh from pointwise.

2. Set up the configuration file, the periodic boundary and grid adaption parts are:

% Periodic boundary marker(s) (NONE = no marker)
% Format: ( periodic marker, donor marker, rot_cen_x, rot_cen_y, rot_cen_z, rot_angle_x-axis, rot_angle_y-axis, rot_angle_z-axis, translation_x, translation_y, translation_z)
MARKER_PERIODIC= ( per1, per2, 0.0, 0.0, 0.0, 0.0, 0.0, 180.0, 0.0, 0.0, 0.0)
% ------------------------- GRID ADAPTATION STRATEGY --------------------------%
%
% Kind of grid adaptation (NONE, PERIODIC)
KIND_ADAPT= PERIODIC

3. Execute the SU2_MSH to output su2 grid with periodic halo element

4. Running the SU2_CFD using new su2 grid from step 3.

In the step 4, it code runs well for serial execution or parallel execution when the core number less than 6. But if I execute the command "SU2_CFD -n 8 periodic.cfd" or use more computational cores than 8, the segmentation error will be occurred at "Establishing MPI communication patterns" or "Setting the control volume structure".

I also try to refine the grid and build more nodes on the periodic boundary surface. In this case, , the segmentation error will be occurred for all parallel executions, no maters how many cores in using . And the serial execution for this refined mesh could runs several steps, however, the segmentation error still be occurred when the master node merge coordinates at solution writing out step.

These are the errors and information I got using 3d rotation periodic BC, and I need you guys' help. Could anyone give me any ideal or suggestion to solve this issue? If anyone try to run my test case, just leave E-mail here and I will sent you the mesh and configuration file.

Thanks!

[stillnash]

I have the same problems with you. Hope you can help me.