Encastre and coupling at the same time. Problem with nodes used twice 1

[OscarPacheco]

Good afternoon to all of you!
I am facing a problem that may be really easy but I can't find the clue.
I have a model of vertebrae C5 and C4 with the intervertebral disc between both.
I want to submit the whole model to a rotation. But I also want to encastre the base of the inferior vertebrae. My aim is to see the resultant moments.
I am using a coupling constraint between a reference point (situated outside the model) and the model itself. I use this reference point to rotate the whole model.
I am using encastre in the base of the vertebrae.
However abaqus exits with an error because nodes used in the coupling are also used in the encastre.
Does anyone know how to solve this problem?

Thank you very much. I really need help.
Please find attached a pdf with images.

 

[testing]

"However abaqus exits with an error because nodes used in the coupling are also used in the encastre."
"Does anyone know how to solve this problem?"

Yes, you don't use nodes in the coupling if they are in the boundary condition.

 

[OscarPacheco]

Thank you zwtipp05 for your answer.
The problem is that I need to submit the whole model to the rotation.
As I need to fix the base, there will be nodes of the base that will be in the boundary condition while they are in the coupling because I have to rotate the whole model.

 

[Mustaine3]

When you have to rotate the whole model, then what is the BC for?

What shall the nodes do that are in both definitions?

 

[OscarPacheco]

Hello Mustaine3.
My aim is to obtain the momentum necessary to rotate the vertebrae a certain angle.
I want to simulate a whiplash. For this, first I need to validate the model.
So imagine the base of the inferior vertebra can't move. Now I want to rotate the rest of the vertebrae and obtain the momentum needed to rotate it.

 

[Mustaine3]

Then make sure that the two selected regions do not overlap. The job will start then.
But I think, that the elements close to the BC will fail pretty soon.

 

[OscarPacheco]

That is the point, the two regions overlap in some nodes. It is obvious because the base is part of the model and is the whole model the one that needs to be submited to rotation.

Do you know another way to do the same but using a different technique?
I have thought about create a new part to act like a table, put the BC to the table, put the model on top of the table, and rotate the model.

 

[Mustaine3]

Again, make sure that these two definitions do not overlap.

When some edges are part of these two regions, then you can easily fix that. Decide in which definition the edge(s) shall not be. Then edit that definition and select again the faces. Before you confirm your selection, press down CTRL key and click on the edges that should not be in there. Confirm that.

 

[IceBreakerSours]

If you are trying to apply a pure moment at the top vertebra so as to replicate a typical pure moment flexibility (ex vivo/cadaveric) test (as per Panjabi and White), then you should couple all nodes at the top vertebra to a reference node (as you have done at the bottom surface of the lower vertebra) and apply a rotation (only) about a given axis at this new reference node.

*********************************************************
Are you new to this forum? If so, please read these FAQs:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[corus]

Remove the encastred nodes from the set when you select the face or whatever. Simply hold down the shift key and pick out the edges or faces that aren't to be coupled.

 

[OscarPacheco]

Thank you all!
Exactly IcebreakerSours! I am trying to replicate Panjabi and White models!
Finally I have opted for the eassiest solution (I didnt think about it before :S ) I am going to put the intervertebral disc below C5 and set the BC there.
However now I have a problem with the contact properties. By now I have tried two options:
• Surface to surface just in the surfaces in touch of different parts; this is interverebral discs with base and tops of vertebrae.
In this case Abaqus exits after 6 attempts with warning messages “Displacement increment for contact is too big”
• General contact. This time applied to the whole model.
In this case Abaqus aborts saying that some of the elements have volume 0.
As you can see the parts are form of multiple instances.
Please can you give me advice in this topic?
Thank you very much! 

 

[IceBreakerSours]

a) If you are replicating a typical ex vivo test, you do NOT fix the disc in an epoxy resin; you fix the vertebral body. So, your easy solution is, in fact, incorrect.
b) Discs do not articulate at the interfaces. You have to tie those nodes.

Be careful! Your basics (FE, functional anatomy of the spine, and ex vivo testing) need some polishing.

*********************************************************
Are you new to this forum? If so, please read these FAQs:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[OscarPacheco]

IceBreakerSours, I am trying to buy the tests of Panjabi but my university biblary is having some problems to get them and I still dont have them...
So acording to you which would be the boundary conditions I would need to apply?
How can I fix the whole vertebral body (as you say in a) and at the same time rotate it?
What do you mean at point b? Should I use as unique interaction a tie between intervertebral discs and vertabrae?
Thank you for your support, I am starting to desperate

 

[IceBreakerSours]

I imagine you meant that you are having trouble getting access to Panjabi and White's book. You could always download a few classic papers (some are freely downloadable) by Panjabi, Goel, Wilke, Patwardhan, and others on bench top testing of spinal functional units. Ideally, you should visit a lab that performs such tests.

A part of the top and the bottom vertebra is fixed in an epoxy/resin mixture so the gripping of either end becomes easy. Bottom end is fixed and the top end is loaded with pure moments, and the motion of each vertebra is computed by post-processing of marker data (with markers rigidly adhered to many locations on the specimen).

*********************************************************
Are you new to this forum? If so, please read these FAQs:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[IceBreakerSours]

Read Vijay Goel, Antonious Rohlmann, and others' papers on finite element modeling of spine.

*********************************************************
Are you new to this forum? If so, please read these FAQs:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083