symmetry planes encore

[AndreMarcic6025]

Hello,

I am working mainly with round parts, so the given boundary conditions limit me to models like half ring or quarter ring.

I try to analyze smaller slices f.e. 20 degree segments. I tried the option to create a new coordinate system, but that was not a big step ahead.

Now it was possible to constrain the slice planes, but if you try to constrain an axial face it is not possible, because the edges of this face have two different CO- systems.

I have added a picture of a typical model, it is like a segment of a pressure vessel.

In Ansys WB there is a constraint called sliding plane(?), more or less similar boundaries can be found in CosmosWorks or Catia.

How can it be done in AD Simulation? (i dont need the oneclick feature like in the other programs, but any work around would be helpful)

regards

Andre

[John_Holtz]

Andre,

I think there are two solutions:

1. Put both faces on the same local coordinate system. That is, a cylindrical coordinate system. The same "symmetry" constraints apply to both faces.
2. do not include the nodes on the centerline on the model in the local coordinate system.

Be sure to read the page "Help > Autodesk Simulation > Setting Up and Performing the Analysis > Set Up Analyses > Local Coordinate Systems". The last figure and the text above it explains why you need to apply a different or additional boundary condition on the nodes along the centerline.

[AndreMarcic6025]

John,

I tried this approach, too

I made a simple ring segment (no centerline nodes).

I created a cylindrical CO system, where the green line (Y) was the theta angle. This boundary has been constrained on the two slice faces and z has been constrained on one axial face. As load I applied a pressure on the inner diameter.

So everything from the setup was fine, attached you see the result.

the second image shows how the CO system is transferred to the surface/nodes and that looks awkward to me. I expected that the small symbols have the same orientation for every node or (better) that they turn with the same angle.

regards

Andre

[John_Holtz]

Andre,

The centerline of the local coordinate system (points A and B in the interface) need to be on the centerline of the ring, not on the lower right corner as indicated in your figure. You may know the coordinates of the centerline axis, or calculate it from the radius and coordinate on the radius, or use construction lines to "find" the centerpoint.