Learning Forum

[m.caragiuli]

Hello,

just to be sure, the data to be inserted in Ansys as tabular values to create a new material and for subsequent curve fitting should be engineering stress/strain values or true stress/strain?

thanks!

[peteroznewman]

Which material model are you asking about?

[m.caragiuli]

Hi,

it is a hyperelastic material the model is Mooney Rivlin, but it could be also Neo-Hookean, as I found experimental data about stress and strain I'd like to know if I should use true strain or engineering one into Ansys.

 

[peteroznewman]

What do the ANSYS Help pages say about experimental data?

[m.caragiuli]

ok, from what I read Ansys requires engineering strain. Thank you!

P.S Could you give me a piece of advise to fix the problem of convergence of my previous post titled "mesh improvement"? I'm at a standstill!

[m.caragiuli]

Hi there,

this is a post about the topic previously discussed, but I'm very confused thus I need a confirmation from you. I've read different answears about the data to be used as input to create a new material from experimental data. Ansys Help reports the use of engineering stress/engineering strain, but about isotropic hardening I've read to input true stress/plastic strain. Why? However, can you confirm to use eng. stress/eng.strain?

Please clarify these aspect. Thanks!

[peteroznewman]

Here are a few excerpts from the ANSYS Help system.

6.3.2. Step 1. Prepare Chaboche Experimental Data

Curve fitting requires experimental test data. To use curve-fitting with plasticity, the only experimental data supported is uniaxial test data. Uniaxial test data has two columns, plastic strain and true stress.

---

6.1.2. Step 1. Prepare Hyperelastic Experimental Data

Table 6.1: Experimental Details for Case 1 and 2 Models and Blatz-Ko

---

So you see that depending on which material model you are building with experimental data, one requires true stress while another requires engineering stress.

[m.caragiuli]

Thank you Peter! It's clear now. 

One more question...if an article shows the figure of a stress-strain curve, how can I understand if it is true stress true strain or eng. stress/eng. strain? 

I have another question about a post processing toot. Not sure if I can ask for it here or on a new post. However, what Ansys means by deformation? Is it the change in shape with respect to the original shape ? If so, which is the difference with the strain?

Thanks

 

[peteroznewman]

The article should define if they are plotting Engineering Stress or True Stress.  You can't tell by looking at a figure.

Deformation is the change in position of a node from its starting point. In a cantilever beam. Nodes at the tip have a large deformation while nodes at the base have a small deformation.

Strain is the change in the length of an element divided by the element length. In a cantilever beam. Elements at the tip have a small strain while elements at the base have a large strain.

[m.caragiuli]

ok, if so is it correct to say that deformation corresponds to the displacement? 

Peter, if I have uniaxial tension and compression data is it correct to use a 2nd order Ogden model to represent the hyperelastic material? I know Ogden is expressed in terms of principal stretches, since uniaxial data involve the stretching and compression along one principal stretch direction I'm not sure if a second order could be applied. It seems that during curve fit solving using a 2nd order ogden, the estimated constants are quite similar to each other. 

Thank you for your tips.

[peteroznewman]

Yes, deformation corresponds to displacement.

I don't know much about elastomer model curve fitting. There are some good articles on this site that give some guidance. Search for them.

I recommend you start a New Discussion to ask about curve-fitting elastomeric data.

[m.caragiuli]

Hi Peter, talking about deformation..looking at the direction deformation (along Y) result of my simulatio, I noticed that there is a displacement along -Y axis (bodies are in the YZ plane Z is vertical, Y is sagittal, but they move anteriorly along -Y and inferiorly along -Z as it should be), but half of the scale values about deformation are positive and half are negative. Why aren't they all neagtive since the movement occurs in the opposite direction of Y axis? Have a look at the picture; I set a 0.5X scale to better appreciate the deformation. Actually it is so small that is not perceivable. (Basically looking at this picture the movement should be from left to right and downwards).

Thanks