Showing results for 
Search instead for 
Did you mean: 

Shell to beam connection


Hi everyone,


I'm trying to study the bottom of a 10m high tubular column. To save computing time I would like to use a shell element for the 2 first meters and then connect a beam element for the 8 last meters till the top.


I think I made mistakes when defining the connection between the top edge of the shell element to the bottom node of the beam because I cannot run the analysis.


Apparently, my "Connections Regions were not defined in topological order."


Can someone help me?



Thanks in advance.






Re: Shell to beam connection

Solution Partner Phenom Solution Partner Phenom
Solution Partner Phenom

Dear Amathias,

Forgot to use the connections edge-to-edge or surface-to-surface, these are useless, here you need to create a spider using RBE2 rigid elements to connect all the nodes of the circular colum (being dependent nodes) with the one-node of the beam (the INDEPENDENT node).






Best regards

Blas Molero Hidalgo, Ingeniero Industrial, Director
Blog Femap-NX Nastran:

Re: Shell to beam connection


Hi, thank you it seems to be exactly what I want to do.


Could you do some printscreen to show me where I can find this function in Femapplease?



Thanks in advance,




Re: Shell to beam connection


Here is my model, I think I have found the function but it doesn't work when I run the analysis.


Can someone help me with that?




Have a nice day,




Re: Shell to beam connection

Siemens Phenom Siemens Phenom
Siemens Phenom

The problem with your model is that the beam is not actually connected to the center node of your RBE2. When I tried to merge the 2 nodes, Femap would not merge them(maybe you tried this?) and the following message gets returned:

Check Coincident Nodes
2 Node(s) Selected...
1 Node Combinations Not Merged to prevent Element corruption
Node Counts do not include Midside Nodes where Corner Nodes were not Merged.
For more information on Merge Failures set the Merge Action to Detailed List
0 Node(s) Merged.


I started with the most common place to make a mistake, the RBE2. What I noticed is that the end node of the beam is actually listed as a dependent node of the RBE2, this is not correct, and that is what is preventing the merge operation. So remove node 4097 from your RBE2 dependent node list, then do a coincident node check. You should see one node merged, now your model is connected and will run expected.


One other note, using the RBE2 means that your tube will remain planar and the tube cross section will remain circular at that location, those nodes have zero relative motion. If you want the tube to ovalize where the beam connects, then you should change to the RBE3.