# [ICEM] How define periodicity correctly in ICEM

 Register Blogs Members List Search Today's Posts Mark Forums Read

 June 13, 2013, 02:55 How define periodicity correctly in ICEM #1 Senior Member   Javi Join Date: Jan 2013 Posts: 276 Rep Power: 9 Hi guys, Iīm dealing with periodicity problems in ICEM. Iīve done "Bottom Up Hexa Mesh Strategy for a Grid Fin" tutorial. Even, when I tried to export this mesh to Fluent I didnīt get it. For that reason, I would like to know how export correctly a periodic mesh to Fluent from ICEM. Iīve tried with a easy example. I followed the same steps as in the tutorial. Before generating the mesh: Mesh -> Global Mesh Set Up -> Set up periodicity. Then, define periodic vertices in the blocks with "Periodic vertices". I convert to unstruct mesh and then I did "Check mesh" and problems related to periodicity appear (I selected the parts with the periodicity)... and then... I donīt know if I have to export the mesh without boundary condition about periodic parts and define them in Fluent... or if I have to define this bc in ICEM... Thank you. I leave this simple geometry if any of you want to try. https://www.dropbox.com/s/bp06ywrnzq1t56m/SetPer.rar

 June 14, 2013, 08:23 #2 Retired from CFD Online     Simon Pereira Join Date: Mar 2009 Location: Ann Arbor, MI Posts: 2,665 Blog Entries: 1 Rep Power: 39 I don't have time to look at your case, but I can tell you the two most common mistakes... 1) When selecting periodic verts, you need to get every pair. A good final check is to right click on faces (in the tree under blocking) and turn on the periodic faces option... You can do the same for Verts and quickly make sure you got it all. 2) The checks ask you to select the parts that are periodic. Often, even if the blocking is periodic, users have associated it to underlying geometry who's names are not setup in periodic pairs. For instance, on the one periodic face, all the surfaces and curves may in in the part PER1, but on the other side all the surfaces are in PER2 and the curves are in GEOM. When you run the periodic check between PER1 and PER2, it won't be a perfect match because you are missing all the line elements and nodes that formed along GEOM... I am not saying that you need all the curves in PER2, but you need to make sure you are comparing equal sets. After that, you do need to do something tricky in Fluent to get it to apply the correct periodic boundary condition... I know I have posted that command several times before, so do a search... Search tip: I find that a google search of the CFD-Online site finds things faster and more reliably than the CFD-Online built in search. ahmadreza likes this. __________________ ----------------------------------------- Please help guide development at ANSYS by filling in these surveys Public ANSYS ICEM CFD Users Survey This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)... CFD Online Users Survey

June 14, 2013, 08:51
#3
Senior Member

Join Date: Feb 2013
Location: Germany
Posts: 189
Rep Power: 16
Quote:
 Originally Posted by PSYMN I don't have time to look at your case, but I can tell you the two most common mistakes... 1) When selecting periodic verts, you need to get every pair. A good final check is to right click on faces (in the tree under blocking) and turn on the periodic faces option... You can do the same for Verts and quickly make sure you got it all.
Especially when you use rotational periodic (didn't look at your case) the vertices on the axxis of rotation should be defined as periodic with themselves. This could also be one common mistake.

 June 14, 2013, 10:15 #4 Retired from CFD Online     Simon Pereira Join Date: Mar 2009 Location: Ann Arbor, MI Posts: 2,665 Blog Entries: 1 Rep Power: 39 @Kad, good one. You need to select those axis verts twice to ensure they are "periodic"... what is really happening is you are defining the corners of each periodic face, so if you forget to select the axis, the faces next to it are not defined as periodic and you are going to have a bad time. __________________ ----------------------------------------- Please help guide development at ANSYS by filling in these surveys Public ANSYS ICEM CFD Users Survey This second one is more general (Gambit, TGrid and ANSYS Meshing users welcome)... CFD Online Users Survey

 June 14, 2013, 10:24 #5 Super Moderator   Sijal Join Date: Mar 2009 Location: Islamabad Posts: 4,358 Blog Entries: 6 Rep Power: 45 Hi Simon greetings Hope you are fine and enjoying good health I have tried everything, but things are not working in ICEM 14.5. Error message:shell xxx has node xxx which has no twin I will come after another try...

 June 14, 2013, 13:13 #6 Super Moderator   Sijal Join Date: Mar 2009 Location: Islamabad Posts: 4,358 Blog Entries: 6 Rep Power: 45 Ok Here is update. From ICEM 14.5 (with full periodicity defined in ICEM) I was able to export the mesh and read into Fluent (see video) and quality and min angle are very good. Quality ~0.9 and min angle >42. However in Fluent min orthogonal quality is 0 (0 is bad and 1 is good and min required is 0.01). Then I have exported mesh from ICEM to Fluent without periodicity and defined periodicity in Fluent (mesh> Modifiy-zone> make-periodic) and now min orthogonal quality is 0.67. cesarcg likes this.

 June 17, 2013, 05:05 #7 Senior Member   Javi Join Date: Jan 2013 Posts: 276 Rep Power: 9 First of all, thanks for the answers. Thank you Simon, the second point is good to know. Exactly, I had the curves in a different part from the "periodic surfaces". Yes, I know how to do this in Fluent. Thank you Kad. This is good one. In pag. 386 Help Manual is also remembered. Hi Far, I have another question very similar at yours. When I check pre-mesh quality in ICEM CFD, for example, Aspect Ratio, I get 12. When I check the Aspect Ratio in Fluent appear 2.45e2. I canīt understand this behaviour. It is two orders of magnitude higher! In addition, I think two criterions are similarly defined. I mention this because Itīs similar to the behaviour you are describing. Thank you guys!

 June 23, 2013, 16:15 tutorial on periodic meshing in ICEM CFD and Fluent #8 Super Moderator   Sijal Join Date: Mar 2009 Location: Islamabad Posts: 4,358 Blog Entries: 6 Rep Power: 45 Here is the link of document https://dl.dropboxusercontent.com/u/...%20meshing.pdf ahmadreza and ama294 like this.

 June 24, 2013, 02:09 #9 Senior Member   Javi Join Date: Jan 2013 Posts: 276 Rep Power: 9 Hi Far, thank you very much for the tutorial!! I did this 1 week ago. In addition, I made 3 video tutorials speaking about different ways to do this but... is in spanish lenguage. Maybe the last one of them (is exactly the way you have explained here) can upload and share the link here as complement of your tutorial.

 June 24, 2013, 04:14 #10 Senior Member   Javi Join Date: Jan 2013 Posts: 276 Rep Power: 9 Here a short video about periodic in Fluent. Make periodic bc in Fluent It just to illustrate farīs tutorial. cesarcg likes this.

 June 24, 2013, 04:46 #11 Senior Member   Javi Join Date: Jan 2013 Posts: 276 Rep Power: 9 Here an other short video about create periodic mesh and set periodic bc in ICEM CFD. Make periodic mesh and bc in ICEM When I check the mesh and appears the window about periodic problems, itīs just because the curves that lie in periodic surface part belong to other part. For more information about that, see Simonīs post (#2). Also, the point I take to base point is created in CAD software and imported to ICEM. I think is easier that way instead of to create it in ICEM.

 June 24, 2013, 06:00 #12 Senior Member   Javi Join Date: Jan 2013 Posts: 276 Rep Power: 9 The las video is about periodic vertices coincidence with base point. Make periodic mesh ICEM Attention from 1:15 to 1:30 m. This vertices have to be periodic too!

June 24, 2013, 06:00
#13
Super Moderator

Sijal
Join Date: Mar 2009
Posts: 4,358
Blog Entries: 6
Rep Power: 45
Quote:
 Originally Posted by FJSJ Here an other short video about create periodic mesh and set periodic bc in ICEM CFD. Make periodic mesh and bc in ICEM When I check the mesh and appears the window about periodic problems, itīs just because the curves that lie in periodic surface part belong to other part. For more information about that, see Simonīs post (#2). Also, the point I take to base point is created in CAD software and imported to ICEM. I think is easier that way instead of to create it in ICEM.
I usually put all curves in part "CRVS" and points in "PTS" and this works very well for hexa and tetra

You can make base point in ICEM. Just take an extreme point (on inlet or outlet) in model and find its coordinates. just remove x and y values and enter Z cooridinate and create new point.

 June 24, 2013, 06:21 #14 Senior Member   Javi Join Date: Jan 2013 Posts: 276 Rep Power: 9 thank you for the advice Far!

 August 28, 2013, 17:24 #15 Member   Cesar Join Date: Nov 2012 Location: Guanajuato, México Posts: 78 Rep Power: 9 Does it always have to be Z the axis of rotation in fluent in order to avoid problems in mesh checking and reporting quality? Last edited by cesarcg; August 28, 2013 at 18:38.

 August 28, 2013, 18:03 #16 Super Moderator   Sijal Join Date: Mar 2009 Location: Islamabad Posts: 4,358 Blog Entries: 6 Rep Power: 45 not necessary PSYMN likes this.

 Thread Tools Display Modes Linear Mode

 Posting Rules You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Smilies are On [IMG] code is On HTML code is OffTrackbacks are On Pingbacks are On Refbacks are On Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post llrr ANSYS Meshing & Geometry 14 February 12, 2017 14:44 Ijaz Fluent UDF and Scheme Programming 4 May 8, 2012 04:24 libia87 Fluent UDF and Scheme Programming 1 May 4, 2012 21:49 Ashi Fluent UDF and Scheme Programming 0 May 25, 2009 09:39 Astrid CFX 6 July 12, 2006 21:46

All times are GMT -4. The time now is 03:51.