CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Siemens (
-   -   Help on cyclic boundary!!! (

Mehdi March 15, 2006 12:37

Help on cyclic boundary!!!

In my problem in cylindrical geometry I have one cell in theta direction and I use cyclic boundaries. In normal grids this boundaries works but when I refine my grids, especially in Z direction I get errors in checking of cyclic boundaries. If you know the solution please help me. I'm thankful for your help in advance.


allan March 16, 2006 07:39

Re: Help on cyclic boundary!!!
try deleting all your cyclic matches: cyde all. Then rematch your cyclic boundary regions with cyma.

Look in prostar help for cyde and cyma.


Mehdi March 16, 2006 09:26

Re: Help on cyclic boundary!!!
Thanks for your comment,

I did it but it doesn't work, after refinement and defining all boundaires, I can pass first checking for boundaries, but in the special check for cyclic boundaries, I get this error message for many of cells:

Two cyclic sets have identical masters, check if they have duplicate slaves or if they can be merged.

If you know, please help me more to solve it

Thanks, Mehdi

allan March 16, 2006 09:47

Re: Help on cyclic boundary!!!
try to put these boundaries into a set, it should do it if you get this error message when writing the geometry file.

check by plotting them

I don't know if there is a command that can merge duplicate cyclic pairs.. perhaps it's something like cymerge, you'll have to check prostar help.

If you have duplicate boundaries, they can be merged after they have been put in a set by:

bmer bset

you may have to redo your cyclic matches again

hope this helps


Mehdi March 16, 2006 12:26

Re: Help on cyclic boundaries(Error 87 and 88) !!!
Thank you very much, I'm beginner and I couldn't understand very good your points, but about merging and redo the cyclic boundaries I did it several times. Yet I couldnot solve it and I always get errors no. 87 and 88.

Tom March 16, 2006 13:23

Re: Help on cyclic boundaries(Error 87 and 88) !!!
I put cyclic boundaries in the same category as cp match errors - works of the devil. One thing to be sure of is that you are using arbitrary rather than integral matches. To merge the boundaries try:

bset,newset,region,XX (where XX is the region number)

bmerge list

Now after saying that, I have never had it work for me. The problem is that if you use a custom mesh, it is quite possible that there is not a one to one match of the faces that you want the cyclis boundary. Therefore, using an arbitrary match is necessary. Also, it may be necessary to zoom in on the refined sections to assure that all of them are chosen in the selection. Sometimes a change in cell size stops the selection if you are using select based on edge angle.

I hope this helped some.


All times are GMT -4. The time now is 21:40.