|
[Sponsors] |
February 7, 2014, 19:43 |
Translational Periodic BC Problem
|
#1 |
New Member
Join Date: Nov 2013
Posts: 5
Rep Power: 13 |
Hello,
I am trying to model a cylinder moving past a stationary cylinder in a flow. To do this I am using a sliding mesh & periodic boundaries but I am running into a problem. I have attached an image of the mesh (the bottom region, between A & B, is the part that should slide to the right). When I try to initialize the solution in Fluent I get errors regarding "Could not intersect threads 11 & 12" (11 & 12 are the IDs of the interface zones, labelled int1 and int2 on my diagram). This is part of my final year project and any help would be much appreciated! My process in Fluent is: 1. Import the mesh from workbench 2. Specify "density based" & "transient" under GENERAL 3. Specify "inviscid" under MODELS 4. Changed zones A & B from walls to translational periodic (using the text UI) 5. "mesh motion" is ticked under CELL ZONE CONDITIONS > UpWind (name of the bottom region in diagram 6. "translational velocity" is specifies as 10m/s in X direction, 0m/s Y direction under CELL NE CONDITIONS > UpWind > Mesh Motion 7. An inlet velocity of 10m/s is specified normal to inlet 8. Outlet BC is pressure outlet 9. An interface is set up under MESH INTERFACES > CREATE. Int1 and Int2 are selected as interface zones and "periodic boundary condition" is ticked under INTERFACE OPTIONS. 10. "translational" is selected as periodic BC type and, as the auto-compute of the offset didn't work, I specified X=10m, Y=0m as the offset. (Assumed that the offset is the distance between the 2 periodic faces) 11. I try to initialize the solution from INLET but that is where I get my errors. Sorry for the long post but hoping that someone can give me an insight into where I'm going wrong. |
|
February 8, 2014, 05:26 |
|
#2 |
Super Moderator
Alex
Join Date: Jun 2012
Location: Germany
Posts: 3,427
Rep Power: 49 |
I think it is step 9 where you went wrong.
The interface option between int1 and int2 should be periodic repeats, not translational periodic. Depending on your setup, you might want to define the left and right faces in the stationary part of the domain as translational periodic, just like you did with faces A and B. I highly recommend you have a look at the sliding mesh totorial before proceeding. BTW: nice post, I wish all questions were this clear. |
|
February 11, 2014, 10:14 |
|
#3 |
New Member
Join Date: Nov 2013
Posts: 5
Rep Power: 13 |
Thank's for the advice! I ended up using the periodic repeats option like you suggested, which required me to make the stationary zone periodic also. This seems to work fine. Much appreciated.
|
|
Tags |
interface, periodic bc, sliding mesh, translational motion |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Periodic and projection problem | crostolo | Pointwise & Gridgen | 9 | March 7, 2017 03:48 |
periodic boundary condition (translational) fluent 6.3 | haihek | FLUENT | 3 | January 22, 2014 05:10 |
[ICEM] how to define periodic in icem ? | mingersai | ANSYS Meshing & Geometry | 1 | February 3, 2012 18:46 |
3Dimensional periodic boundary condition bench mark problem for incompressible | HaKu | Main CFD Forum | 0 | September 7, 2011 14:48 |
periodic flow problem | Alok Gupta | FLUENT | 1 | November 19, 2003 08:05 |