Strange Beam Behavior

[Joseph Kim]

Hello all,

I am trying to model a cuboct lattice geometry using beam sections in ABAQUS.

I have noticed some strange behavior when modeling the geometry, depending on how I define my beam sections.

Each of my subunits that is used to build this lattice contains 6 nodes and 12 beams. I am setting the beam orientation of these 12 beams by grouping them into 12 different beam sections. I've attached a picture of the part and lattice for reference.

Strangely, I've noticed different behavior (stress distribution and maximum stress) depending on whether I split the beam section definitions into 12 or group them all into a single beam section, even if the parameters for the beam section is identical. This is very strange behavior. I thought maybe it was due to nodal normal averaging occurring, but none of the normals are within the 20 degree range necessary to trigger this behavior in ABAQUS.

I've attached two different INP files for reference, one with a single beam section definition and one with them split into 12 separate beam sections, but with identical parameters.

I would appreciate any insight into what might be causing this.

Thank you for your help.

1 beam section:

*Heading
Default Title
*Preprint, echo=NO, model=NO, history=NO, contact=NO
**
** PARTS
**
*Part, name=beam
*Node
1, -0.14564753, 0.0, 0.0
2, 0.0, -0.14564753, 0.0
3, 0.14564753, 0.0, 0.0
4, 0.0, 0.14564753, 0.0
5, 0.0, 0.0, 0.14564753
6, 0.0, 0.0, -0.14564753
8, 0.29129506, -0.14564753, 0.0
9, 0.43694259, 0.0, 0.0
10, 0.29129506, 0.14564753, 0.0
11, 0.29129506, 0.0, 0.14564753
7, 0.29129506, 0.0, -0.14564753
*Element, type=B33
1, 1, 2
2, 2, 3
3, 3, 4
4, 4, 1
5, 1, 5
6, 2, 5
7, 3, 5
8, 4, 5
9, 1, 6
10, 2, 6
11, 3, 6
12, 4, 6
13, 3, 8
14, 8, 9
15, 9, 10
16, 10, 3
17, 3, 11
18, 8, 11
19, 9, 11
20, 10, 11
21, 3, 7
22, 8, 7
23, 9, 7
24, 10, 7
*Elset, elset=Ultem2200_SECOND_001
2, 14
*Elset, elset=Ultem2200_FOURTH_001
4, 16
*Elset, elset=Ultem2200_UPPER_FIRST_001
5, 17
*Elset, elset=Ultem2200_UPPER_FOURTH_001
8, 20
*Elset, elset=Ultem2200_LOWER_FIRST_001
9, 21
*Elset, elset=Ultem2200_LOWER_SECOND_001
10, 22
*Elset, elset=Ultem2200_FIRST_001
1, 13
*Elset, elset=Ultem2200_UPPER_SECOND_001
6, 18
*Elset, elset=Ultem2200_LOWER_FOURTH_001
12, 24
*Elset, elset=Ultem2200_THIRD_001
3, 15
*Elset, elset=Ultem2200_UPPER_THIRD_001
7, 19
*Elset, elset=Ultem2200_LOWER_THIRD_001
11, 23
*Elset, elset=Ultem2200
1, 2, 3, 4, 5, 6, 7, 8, 9, 10
11, 12, 13, 14, 15, 16, 17, 18, 19, 20
21, 22, 23, 24
*Orientation, name=Ori-1
1, 0, 0, 0, 1, 0
1, 0
*Beam Section, elset=Ultem2200, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*End Part
*Assembly, name=Assembly
*Instance, name=beam-1, part=beam
*End Instance
*Nset, nset=fixed_nodes, instance=beam-1
6, 7
*Nset, nset=moved_nodes, instance=beam-1
5, 11
*Elset, elset=all_elements, instance=beam-1
1, 2, 3, 4, 5, 6, 7, 8, 9, 10
11, 12, 13, 14, 15, 16, 17, 18, 19, 20
21, 22, 23, 24
*End Assembly
** MATERIALS
** 
*Material, name=Ultem2200
*Elastic
2000000000, 0.3
*Density
0.0001329
** 
** BOUNDARY CONDITIONS
** 
** Name: BC-1 Type: Symmetry/Antisymmetry/Encastre
*Boundary
**pinned_nodes, 1, 3, 0
fixed_nodes, ENCASTRE
*Step, name = Step-1, nlgeom=No
*Static
1., 1., 1e-05, 1.
** 
** LOADS
** 
** Name: point_load   Type: enforced displacement
*Cload
moved_nodes, 3, -10
** 
** OUTPUT REQUESTS
** 
*Restart, write, frequency=1
** 
** FIELD OUTPUT: F-Output-1
** 
*Output, field, variable=PRESELECT
** 
** HISTORY OUTPUT: H-Output-1
** 
*Output, history, variable=PRESELECT
** 
*El print, elset=all_elements
MISES
*End Step

12 sections:

*Heading
Default Title
*Preprint, echo=NO, model=NO, history=NO, contact=NO
**
** PARTS
**
*Part, name=beam
*Node
1, -0.14564753, 0.0, 0.0
2, 0.0, -0.14564753, 0.0
3, 0.14564753, 0.0, 0.0
4, 0.0, 0.14564753, 0.0
5, 0.0, 0.0, 0.14564753
6, 0.0, 0.0, -0.14564753
8, 0.29129506, -0.14564753, 0.0
9, 0.43694259, 0.0, 0.0
10, 0.29129506, 0.14564753, 0.0
11, 0.29129506, 0.0, 0.14564753
7, 0.29129506, 0.0, -0.14564753
*Element, type=B33
1, 1, 2
2, 2, 3
3, 3, 4
4, 4, 1
5, 1, 5
6, 2, 5
7, 3, 5
8, 4, 5
9, 1, 6
10, 2, 6
11, 3, 6
12, 4, 6
13, 3, 8
14, 8, 9
15, 9, 10
16, 10, 3
17, 3, 11
18, 8, 11
19, 9, 11
20, 10, 11
21, 3, 7
22, 8, 7
23, 9, 7
24, 10, 7
*Elset, elset=Ultem2200_SECOND_001
2, 14
*Elset, elset=Ultem2200_FOURTH_001
4, 16
*Elset, elset=Ultem2200_UPPER_FIRST_001
5, 17
*Elset, elset=Ultem2200_UPPER_FOURTH_001
8, 20
*Elset, elset=Ultem2200_LOWER_FIRST_001
9, 21
*Elset, elset=Ultem2200_LOWER_SECOND_001
10, 22
*Elset, elset=Ultem2200_FIRST_001
1, 13
*Elset, elset=Ultem2200_UPPER_SECOND_001
6, 18
*Elset, elset=Ultem2200_LOWER_FOURTH_001
12, 24
*Elset, elset=Ultem2200_THIRD_001
3, 15
*Elset, elset=Ultem2200_UPPER_THIRD_001
7, 19
*Elset, elset=Ultem2200_LOWER_THIRD_001
11, 23
*Elset, elset=Ultem2200
1, 2, 3, 4, 5, 6, 7, 8, 9, 10
11, 12, 13, 14, 15, 16, 17, 18, 19, 20
21, 22, 23, 24
*Orientation, name=Ori-1
1, 0, 0, 0, 1, 0
1, 0
*Beam Section, elset=Ultem2200_SECOND_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*Beam Section, elset=Ultem2200_FOURTH_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*Beam Section, elset=Ultem2200_UPPER_FIRST_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*Beam Section, elset=Ultem2200_UPPER_FOURTH_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*Beam Section, elset=Ultem2200_LOWER_FIRST_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*Beam Section, elset=Ultem2200_LOWER_SECOND_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*Beam Section, elset=Ultem2200_FIRST_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*Beam Section, elset=Ultem2200_UPPER_SECOND_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*Beam Section, elset=Ultem2200_LOWER_FOURTH_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*Beam Section, elset=Ultem2200_THIRD_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*Beam Section, elset=Ultem2200_UPPER_THIRD_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*Beam Section, elset=Ultem2200_LOWER_THIRD_001, material=Ultem2200, temperature=GRADIENTS, section=RECT
0.01, 0.01
0, 0, 1
*End Part
*Assembly, name=Assembly
*Instance, name=beam-1, part=beam
*End Instance
*Nset, nset=fixed_nodes, instance=beam-1
6, 7
*Nset, nset=moved_nodes, instance=beam-1
5, 11
*Elset, elset=all_elements, instance=beam-1
1, 2, 3, 4, 5, 6, 7, 8, 9, 10
11, 12, 13, 14, 15, 16, 17, 18, 19, 20
21, 22, 23, 24
*End Assembly
** MATERIALS
** 
*Material, name=Ultem2200
*Elastic
2000000000, 0.3
*Density
0.0001329
** 
** BOUNDARY CONDITIONS
** 
** Name: BC-1 Type: Symmetry/Antisymmetry/Encastre
*Boundary
**pinned_nodes, 1, 3, 0
fixed_nodes, ENCASTRE
*Step, name = Step-1, nlgeom=No
*Static
1., 1., 1e-05, 1.
** 
** LOADS
** 
** Name: point_load   Type: enforced displacement
*Cload
moved_nodes, 3, -10
** 
** OUTPUT REQUESTS
** 
*Restart, write, frequency=1
** 
** FIELD OUTPUT: F-Output-1
** 
*Output, field, variable=PRESELECT
** 
** HISTORY OUTPUT: H-Output-1
** 
*Output, history, variable=PRESELECT
** 
*El print, elset=all_elements
MISES
*End Step

 

[Igor76]

Hi
please insert picture for second case

 

[Joseph Kim]

Hi,

So the pictures of the beam orientations look identical for either case (as you would expect because the beam section definitions are identical).

That's why I'm thinking it has something to do with the nodal normals that is somehow introducing some twist into the beams. Unfortunately, I don't think ABAQUS has an option to plot those beam normals because they can vary per element.

I'm getting this from the beam section orientation in the ABAQUS manual here:

http://50.16.225.63/v2016/books/usb...06ch29s03alm09.html#usb-elm-ebeamcrosssection

 

[Joseph Kim]

I've attached two pictures to this post that shows that the stress distribution and maximum stress do vary depending on whether you split the beam section definitions into 12 identical or just 1.

12 beam sections:

1 beam section:

 

[Igor76]

Hi
I reassign beam section orientation to 0,0,-1 for both models and have got nearly the same results for both models.You have to check tangent directions also.

 

[Joseph Kim]

Hello Igor,

Thank you for trying to assist me with this issue. So the input files are generated directly by custom software, so the GUI is only used for viewing the results. This means that the tangent directions are always the same for the two files (you can see that the node and beam element definitions in the two files are identical).

I tried changing the beam orientation to 0, 0, -1 for both models and got the same results as above. The model is completely symmetric so I would not expect changing the beam orientation from +Z to -Z to change anything about the model.

I have already reached out to the ABAQUS support team for some assistance, but they have always been notoriously slow in getting back to me.

I would appreciate any other ideas about what is going on.

Igor, could you post the two input files that allowed you to get the nearly identical results for both models? I will go through and try to debug the issue.

 

[Igor76]

Hi
Here attached I`m sending you two input files.

 

[Joseph Kim]

Just as follow up in case people were interested,

I did get in contact with ABAQUS support team and this has been confirmed to be a bug. I will give another update once it is resolved

Thanks for your help!

 

[Igor76]

Hi Joseph
Thank you for information.If its possible,ask them please,why lineload in local (beam) cs has only 2 components.And second question: why assy.ips.pointOn return coordinates of yellow points on edges in part cs (not assy cs) in contrast with assy.edge.pointOn behaviour.Thank you in advance,unfortunately I dont have ability to contact with support team.I'm sure this information will be interesting for you too.

 

[Joseph Kim]

Sure, I will try my best to relay the questions. They are presumably busy during the holidays.