Relationship between stress , rigid force and reaction forces

[athpa]

Dear all,

I am trying to simulate scaffolds of water based matrix embedded with fibers in different angles (15, 30, 45 respectively). I worked with materials (holzapfel-gasser-ogden uncoupled, uncoupled solid mixture) but my results differ a lot with the expeirimental especially the rigid force/reaction forces. Firstly I would like to ask if there is any relationship between the stresses and rigid force in post process. For example I found in the repository a model called uniaxial_mixture.fs2. This model captures the essential difference by changing angles during the tensile test. What I don't understand is how the forces are so much greater than the stresses. Is it mentioned in the users manual ?

image.pngimage.png

Secondly in my model which you can find it attached uniaxial_mixture.fs2 , I am still receiving values that do not make sense with the reality. For this I have to ask the following questions :

a) Do you recommend any material which take into consideration only the E modulus of the matrix and not the fibers ? Because what I suspect is that the solid mixture takes into account the modulus of fibers, which is really high (~1200 MPa). However in the reality the water based matrix (~5 MPa) affects mostly the experimental results of the scaffold (stress around 2.5 MPa, while Forces around 5 N).

b) Do you suggest to change the mesh ? https://forums.febio.org/forum/febio...ariable-filter

c) In the attached file scaffold_hgo in order to run the model, I had to increase c from 2.5 (real value) to 10 ! and then by changing the angle from 5 to 10, rigid force remains the same which doesn't make sense.


Best regards,
athpa

[maas]

Hi Athpa,

Regarding the uniaxial_mixture.fs2 file, I don't see any issues here. You say that the force seems much bigger than the stress, which is true, but you are not comparing apples to apples, since these quantities are in different units. Even more, if you calculate the integral of the Y-stress on the top surface of the tissue you get 68, and the Y component of the rigid force is 70, so I think everything is fine. (The small discrepancy is likely due to the averaging of the stress over the element when febio writes the plot file.)

I cannot run the scaffold.fs2 file. It error terminates with nans.

The scallfold_hgo.fs2 file works fine and I do see a change in the reaction force when I change to angle (gamma) from 5 to 10, so I cannot reproduce your findings.

Regarding your questions:

a) If you set the k1 and k2 parameters to zero, the fibers should not contribute, and you would only get the matrix response.
b) you could try a finer mesh, but I don't think it will make a big difference. Since this model runs fast enough, it's definitely something to try.
c) As I mentioned, I cannot reproduce this.

I hope this helps. Let us know if you have any further questions.

Best,

Steve

[athpa]

Hi Athpa,

Regarding the uniaxial_mixture.fs2 file, I don't see any issues here. You say that the force seems much bigger than the stress, which is true, but you are not comparing apples to apples, since these quantities are in different units. Even more, if you calculate the integral of the Y-stress on the top surface of the tissue you get 68, and the Y component of the rigid force is 70, so I think everything is fine. (The small discrepancy is likely due to the averaging of the stress over the element when febio writes the plot file.)

I cannot run the scaffold.fs2 file. It error terminates with nans.

The scallfold_hgo.fs2 file works fine and I do see a change in the reaction force when I change to angle (gamma) from 5 to 10, so I cannot reproduce your findings.

Regarding your questions:

a) If you set the k1 and k2 parameters to zero, the fibers should not contribute, and you would only get the matrix response.
b) you could try a finer mesh, but I don't think it will make a big difference. Since this model runs fast enough, it's definitely something to try.
c) As I mentioned, I cannot reproduce this.

I hope this helps. Let us know if you have any further questions.

Best,

Steve

Thank you Steve. I will have a deep look again and I ll write again.

Btw with the new version of FEBio and FEBio Studio a new parameter appeared (mu) in material menu. But that parameter is not documented in user manual. Is it by mistake ?


image.png


Best,
athpa

[maas]

Hi Athpa,

Regarding the "mu" parameter, it looks like this adds a shearing contribution to the fiber stress. I am not sure why this parameter was added, but I suspect you can just ignore it for most applications.

Best,

Steve