Help needed! 4 point bending on Bone Constructs

[ka46]

Hi!!

I am currently running static analysis of bone constructs with a bone plates attached. I have done axial loading static analysis and results were in agreement with the experiments that my team has carried out. However, using the same model, the results by Abaqus deviates significantly from that of experiment. We wanted to study the displacements and stresses experienced by the system with the use of the bone plate to repair bone fractures. Experiments were verified to be in accordance to relevant procedures.

Would like to ask whether if the following loading/boundary conditions and contact types that I have used are correct.

Link to download pictures of my model
Link


Contact used:
[ul]
[li]I have split the model into half and re-merged them in Abaqus. I merged the screw heads with the holes of the bone plate to simulate bonded contact with the screw plate.[/li]
[li]I have also merged the bottom of the screws with the far side of the bone to simulate bonded contact.[/li]
[li]The bone parts are merged to simulate them as a single half[/li]
[li]The near side where the middle section of screw is in contact with the holes drilled in the bone, Hard Contact was used to simulate non-bonded contact[/li]
[li]Tangential Frictionless contact between the plate and bone surface as friction is considered to be insignificant[/li]
[/ul]

Constraints:
[ul]
[li]I have used Tie Constraints to tie the top and bottom half of the screws as well as the bone so as to simulate them as a single piece.[/li]
[/ul]

Force and Boundary Conditions
[ul]
[li]For Axial loading, I restrained one side in all directions and applied concentrated load on the other side[/li]
[li]For 4 point bending, I restrained one support (3 nodes) in all directions and the other support (3 nodes) in Y and Z direction. The concentrated loads were applied at 2 different locations (3 nodes each) marked by the boundaries on the model.[/li]
[/ul]

Mesh
[ul]
[li]A mesh of seed size 5 were used. Quadratic Tet was used throughout the model.[/li]
[/ul]


Thanks in advance for any comments on this.

 

[IceBreakerSours]

I have done axial loading static analysis and results were in agreement with the experiments that my team has carried out. However, using the same model, the results by Abaqus deviates significantly from that of experiment.

What's wrong again?

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

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

 

[ka46]

Hi.

We are creating an FE model that replicates the experiment so that the model can be used for further simulations for other conditions that can be applied to the bone construct.

The 4 point bending results from the experiment produced a much higher displacement as compared to that produced by the FE model.

 

[IceBreakerSours]

Assuming you are: a) consistent in units, b) applying correct amount of load, c) using appropriate material models & parameters, d) running the correct type of analysis, and e) measuring the displacement correctly, I can't think of anything obvious. However, I'd try increasing the number of elements in the plate along its thickness direction.

If you wish, uploading your INP will allow us to look at the model in more detail.

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

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

 

[burningDiesel]

Was the axial loading done both in tension and compression?
Mesh for the plate seems coarse, while for displacements it can be ok, even in bending, I would doublecheck it (refine it).

 

[corus]

Tetrahedral elements are ok in axial loading but rubbish in bending, tending to be over stiff. Spend a little more time on the meshing and use brick elements and you should find much better results.

 

[IceBreakerSours]

Click here and watch 8:30 onwards for more on tetrahedral elements.

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

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

 

[ka46]

Thanks for the comments. Will give it a try.