Cyclic symmetry- Number of segments to output

disla

Hello,

I’m not sure to understand properly the meaning of the “Number of segments to output=All” inside the cyclic symmetry BC.
Once solved, only the seed sector is shown in the screen.
¿How can I get all sectors represented to check any strange behavior? , like the Impeller example in the manual where the whole Impeller is visualized although only one blade is computed?

Thanks

Sergio

In case of using CCX solver, check the generated input file and look for this line. N is the number of sectors that makes the whole goemetry, and NGRAPH is the quantity that will be saved on the results. For having a complete representation both numbers must be the same.

*CYCLIC SYMMETRY MODEL, N=12, NGRAPH=3
0.,0.,0.,1.,0.,0.

disla

Hello Sergio,

Cyclic symmetry seems not allowed in Mecway with ccx. It turns red.
The inp file generated by Mecway does not contain CYCLIC SYMMETRY card.
If I import the previously "save as inp file" CYCLIC SYMMETRY seems that has disappeared.
¿Am I doing something wrong?

disla

Sorry I forgot to comment I'm using MECWAY 9 and ccx_2.14_MT.exe.

Sergio

Now I'm not remember well if is included or not in the Mecway interface, but you can add the card by means of custom CCX data on the tree. Check the attached model

Sergio

By the way, the actual sector has a bad angle, about 9.xx degrees, so there will be more than 36 sectors. That's why in the results there is a small gap between sector, and I was forced to use the CHECK=NO option to be able to solve it anyway.

Victor

With the internal solver, the number of output segments is only used for modal vibration. In other analysis types, the results are identical for each segment so it would mostly be cosmetic to show them. I agree that this should though, it's just missing functionality.

In Mecway Version 10, cyclic symmetry will work directly with CCX.

disla

I really did not pay very much attention to the angle.
That is a small part of the complete dome with its supporting structure. It helps me to understand properly what am i doing before going to the complex geometry and waste too much time.

¿ Does it have sense to provide a number of sectors that do not complete the 360º ?
If I indicate 4 sectors they are "disconected" one from the other.
Seems there should be no need to ask for them. ¿Isn't it?

Victor

They have to add up to 360 degrees or the solution will be wrong (internal solver) or fail (CCX). There is some redundancy there and in principle, it could estimate the number of segments from the mesh. But there may be ambiguous cases like the nodes on the cylindrical surfaces in the attached picture.

disla

Thank you very much to both for your help.