asdeasoft.net

Thank you for clarification - makes sense. In my case I would select middle option (w rigid link) given my modeling approach.

I have a few follow up questions:
1) How do I implement this in my big building model? In my understanding I need to split geometries of the walls corresponding to beam depths (see red line in the screenshot) and the apply rigid links (see blue lines in the screenshot). Can you please confirm that this is how you do it?
2) If my understanding in #1 is correct, what is the most effective way to do this without redoing the entire wall geometry? Is there a comment where I can simply split existing geometries?
3) regarding application of rigid links, since i have many coupling beams, do I need to create many Interactions and many Conditions, or I can get away with using only one Interaction and one Condition?

Please let me know. Thank you.

If you want to use the RigidLink approach:

1) How do I implement this in my big building model? In my understanding I need to split geometries of the walls corresponding to beam depths (see red line in the screenshot) and the apply rigid links (see blue lines in the screenshot). Can you please confirm that this is how you do it?

Yes

2) If my understanding in #1 is correct, what is the most effective way to do this without redoing the entire wall geometry? Is there a comment where I can simply split existing geometries?

You can draw lines that pass through the faces at the level of each beam's depth, then use the Merge command. It will keep materials/element/conditions assignments

3) regarding application of rigid links, since i have many coupling beams, do I need to create many Interactions and many Conditions, or I can get away with using only one Interaction and one Condition?

No, since the condition type is the same, you can simply create 1 interaction with 1 condition


However, there is another effective approach that I forgot to mention in the previous post. And it's also much easier to implement because you don't need to do any changes to your building model.
You can use the ASDEmbeddedNodeElement condition to tie the beam-ends nodes to the walls.

There is only one small change you have to do in your model: The wall's mesh size should be comparable to the beam depth.
If you do this, you can create interactions to "embed" the beam node to the nearest wall element, which now has about the same size as the beam depth.
The MASTER is the wall face, and the SLAVE is the beam-end node. STKO will automatically find the nearest mesh-element (That's why it is important that the mesh size is close to the beam size, because only 1 element will be connected to the beam)
The Penalty parameter of the Embed condition should be:
P = G*thickness*100 (i.e. The shear modulus times the wall's thickness, increased by 2 orders of magnitude, just to make this stiffer than the real shear stiffness of the wall)

This is a modified version of the previous test, where the third option is the Embedded one.
With this approach, the walls and beam can be attached in the same geometry (as in your case) or detached. It's the same. In this example I kept them connected because it's the way that minimizes the changes to your model
When you apply this method to your multi-story building you just have to pay attention to this:
STKO will automatically find, for each slave node, the closest wall's mesh element.
When a beam is at the intersection between 2 floors, STKO can randomly pick the element above or below, because they have the distance from the beam's end-node.
However, in your case you want to connect them to the floor below.
So you can simply follow the approach in this file: Create 2 interactions, one for alternate floors. in such a way that, in each interaction, the beam can only be attached to the wall, below, because the wall above will be in the other interaction.
Then you can simply create 1 condition for both interactions

Thank you very much for all these examples and explanations. This is really useful.

I made model with rigid links and it seems to be working OK - still need to check results in details.

Out of curiosity I also made a model with embedding the beam into the wall (see screenshot below), but I get the some kind of error (see screenshot below) and the analysis doesnt even start to run. Any ideas why is this happening?

I made model with rigid links and it seems to be working OK - still need to check results in details.

Ok

Out of curiosity I also made a model with embedding the beam into the wall (see screenshot below), but I get the some kind of error (see screenshot below) and the analysis doesnt even start to run. Any ideas why is this happening?

There is a strange segmentation fault. Are you using it with OpenSeesMP?
Are you using the latest version from here? https://drive.google.com/file/d/1ETGQ3Y ... sp=sharing
Try downloading the latest version an let me know.
If it doesn't work, could you please send me the model? If you cannot share it you can send it to me vai e-mail:
[email protected]