Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Piecewise Function to Define Young's modulus

Please login with a confirmed email address before reporting spam

Hi everyone,

I tested some samples of a new material (tension & compression test) in order to obtain mateiral properties (at first Young's modulus). I want to compute a Comsol simulation with this material, so I have to define its behaviour which is non lineare. In particular, for strain>0 (tension) it shows a linear behaviour, while for strain<0 its behaviour is bilinear, so we have one value for young's modulus corresponding to small strain, and another one corresponding to large strain.

I though that one possible solution to describe this could be a piecewise function through which define the Elastic Modulus of this material (E=pw1) for example, but I have some doubts.

Could you help me please?

Maria


7 Replies Last Post 29.10.2019, 07:05 GMT-4
Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 28.10.2019, 06:22 GMT-4
Updated: 5 years ago 28.10.2019, 02:22 GMT-4

Hi,

What you are looking for is probably the Nonlinear Elastic Material with the input option Uniaxial data.

Regards,
Henrik

-------------------
Henrik Sönnerlind
COMSOL
Hi, What you are looking for is probably the **Nonlinear Elastic Material** with the input option **Uniaxial data**. Regards, Henrik

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 28.10.2019, 07:52 GMT-4

Hi Henric,

thank you so much for your reply. I tried to use it, but the fact is that I have a lot of samples tested, each with different behaviour, so I don't understand how I can define a large number of curves which describe the stress-strain relationship.

This is my doubt. Could you help me?

Thanking you in advance, BR

Maria

Hi Henric, thank you so much for your reply. I tried to use it, but the fact is that I have a lot of samples tested, each with different behaviour, so I don't understand how I can define a large number of curves which describe the stress-strain relationship. This is my doubt. Could you help me? Thanking you in advance, BR Maria

Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 29.10.2019, 03:55 GMT-4
Updated: 5 years ago 28.10.2019, 23:55 GMT-4

Hi Maria,

I do not really understand the question here. Are you interested in some kind of averaging of the measured curves, or do you want to make several analyses, using different curves?

Regards,
Henrik

-------------------
Henrik Sönnerlind
COMSOL
Hi Maria, I do not really understand the question here. Are you interested in some kind of averaging of the measured curves, or do you want to make several analyses, using different curves? Regards, Henrik

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 29.10.2019, 06:10 GMT-4

Hi Enric,

the fact is that I have tested several samples of the same material, having different mechanical properties. In particular, during the compression test, they show different behaviour, so the curve stress-strain of the first sample, for example, changes its slope at one specific value of strain, while the curve corresponding to another sample changes its slope at different strain value. My issue derives from these different behaviour. I'm interested of averaging of the measured curves, but it seems difficult to do this.

I would like to ask you if I could define a Piecewise Function where Young Modulus is defined with refers to strain, so that i could define different strain intervals where Young Modulus assumes different values. Then, under the node "Linear Elastic Material", I could use this function in order to define the mechanical properties of the material.

Is it a good approach?

Thanking you in advance, BR

Maria

Hi Enric, the fact is that I have tested several samples of the same material, having different mechanical properties. In particular, during the compression test, they show different behaviour, so the curve stress-strain of the first sample, for example, changes its slope at one specific value of strain, while the curve corresponding to another sample changes its slope at different strain value. My issue derives from these different behaviour. I'm interested of averaging of the measured curves, but it seems difficult to do this. I would like to ask you if I could define a Piecewise Function where Young Modulus is defined with refers to strain, so that i could define different strain intervals where Young Modulus assumes different values. Then, under the node "Linear Elastic Material", I could use this function in order to define the mechanical properties of the material. Is it a good approach? Thanking you in advance, BR Maria

Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 29.10.2019, 06:25 GMT-4

Hi Maria,

looking at your question from a physical point of view, I don't think averaging material properties with large variations is a good approach. I would rather think that you need to run the simulations using different material curves in order to see which way they affect the simulation result you are interested in.

This may be particularly critical as the model seems to be nonlinear.

My two cents. Cheers Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Hi Maria, looking at your question from a physical point of view, I don't think averaging material properties with large variations is a good approach. I would rather think that you need to run the simulations using different material curves in order to see which way they affect the simulation result you are interested in. This may be particularly critical as the model seems to be nonlinear. My two cents. Cheers Edgar

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 29.10.2019, 06:52 GMT-4

Hi Edgar, thank you for your reply.

I agree with you, but I cannot compute several simulations. For my work, I need to find a single simulation that can accurately describe the material behaviour.

Now I'm trying to define Young modulus E wrt strain (solid.eZZ). Do you think this could be a valid approach?

Thanking you in advance, BR

Maria

Hi Edgar, thank you for your reply. I agree with you, but I cannot compute several simulations. For my work, I need to find a single simulation that can accurately describe the material behaviour. Now I'm trying to define Young modulus E wrt strain (solid.eZZ). Do you think this could be a valid approach? Thanking you in advance, BR Maria

Edgar J. Kaiser Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 29.10.2019, 07:05 GMT-4

Maria,

the model may then be correct for this specific material curve. However, with just one simulation you won't be able to validate the model for variations in the material behavior and this seems to be a critical part in your task.

So, no, it is probably not a valid approach.

Cheers Edgar

-------------------
Edgar J. Kaiser
emPhys Physical Technology
www.emphys.com
Maria, the model may then be correct for this specific material curve. However, with just one simulation you won't be able to validate the model for variations in the material behavior and this seems to be a critical part in your task. So, no, it is probably not a valid approach. Cheers Edgar

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.