Menu

Discussion Forum

New Discussion
Filtrer

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.

How to define a new Anisotropic Creep Law?

Posted 13 oct. 2023, 09:27 UTC−4 General, Structural Mechanics, Equation-Based Modeling Version 6.0 5 Replies

Shyam Kishor
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi all, I want to implement a new creep law in to find the creep behaviour of a pressure tube. The expression for the axial creep rate and transverse creep rate are different and given as,

eps_dot_axial= (E1*sigma_t - E2*sigma_a)*K1*phi +G1*K2*phi

eps_dot_transverse= (E3*sigma_t - E4*sigma_a)*K1*phi +G1*K2*phi

where sigma_t and sigma_a are the transverse and axial stress inside the pressure tube and all others are constants and well known.

How can I define this in COMSOL to get the ansotropic creep strain of the pressure tube?

Thanks for your suggestions.

5 Replies Last Post 21 oct. 2023, 01:05 UTC−4
Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 year ago 16 oct. 2023, 03:43 UTC−4

The User defined option in the Creep node allows only isotropic creep. You do however have a possibility to define a general anisotropic creep law by adding an Inelastic Strain Rate node.

If you are working in 3D, select a cylindrical coordinate system. Use the same coordinate system in the parent Linear Elastic Material node, so that you have access to the corresponding stress components.

If you are working in a 2D axisymmetric geometry, the cylindrical system is part of the formulation.

-------------------
Henrik Sönnerlind
COMSOL
The *User defined* option in the *Creep* node allows only isotropic creep. You do however have a possibility to define a general anisotropic creep law by adding an *Inelastic Strain Rate* node. If you are working in 3D, select a cylindrical coordinate system. Use the same coordinate system in the parent *Linear Elastic Material* node, so that you have access to the corresponding stress components. If you are working in a 2D axisymmetric geometry, the cylindrical system is part of the formulation.
Shyam Kishor
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 year ago 17 oct. 2023, 08:22 UTC−4

Thank you Henrik for your response. Your suggestion will help me a lot. However I am new to use inelastic strain rate node. Can you please give me more explanations on how to use inelastic strain rate node to define this creep law. Thank you very much for suggestions.

Thank you Henrik for your response. Your suggestion will help me a lot. However I am new to use inelastic strain rate node. Can you please give me more explanations on how to use inelastic strain rate node to define this creep law. Thank you very much for suggestions.
Henrik Sönnerlind COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 1 year ago 18 oct. 2023, 02:24 UTC−4

The entries are the components of the strain rate tensor. Essentially, you just have to select the format of the tensor (Diagonal or Symmetric) and then use the expressions from your first posting, but substituted with whatever the quantites would be called in your model. The stresses would be something like

sigma_t -> solid.s22 sigma_a -> solid.s33

Note that you need to provide the full tensor (even though you may decide to set some components to zero, assuming that there is no creep in that direction).

An important aspect is whether the physical creep process is volume preserving or not. If it is, the strain rate tensor must be designed to have that property. For example, the radial strain rate (not mentioned by you), cannot be chosen arbitrarily, since in this case it is required that trace(epsilon_dot) = 0.

-------------------
Henrik Sönnerlind
COMSOL
The entries are the components of the strain rate tensor. Essentially, you just have to select the format of the tensor (Diagonal or Symmetric) and then use the expressions from your first posting, but substituted with whatever the quantites would be called in your model. The stresses would be something like sigma_t -> solid.s22 sigma_a -> solid.s33 Note that you need to provide the full tensor (even though you may decide to set some components to zero, assuming that there is no creep in that direction). An important aspect is whether the physical creep process is volume preserving or not. If it is, the strain rate tensor must be designed to have that property. For example, the radial strain rate (not mentioned by you), cannot be chosen arbitrarily, since in this case it is required that trace(epsilon_dot) = 0.
Shyam Kishor
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 year ago 20 oct. 2023, 00:56 UTC−4

Thanks, Henric for your response.I will try it.

Thanks, Henric for your response.I will try it.
Shyam Kishor
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 year ago 21 oct. 2023, 01:05 UTC−4

Hi Henric, First I described the creep rate equation in the variable subnode. Then use it in the inelastic strain rate tensor, But I am facing many convergence errors. Have a look on my model which I attached. Thanks for your suggestions.

Hi Henric, First I described the creep rate equation in the variable subnode. Then use it in the inelastic strain rate tensor, But I am facing many convergence errors. Have a look on my model which I attached. Thanks for your suggestions.

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.

Suggested Content