Implementation of a creep-damage model in ABAQUS using subroutines

[ttavio]

Hello,

I am trying to implement a subroutine in ABAQUS regarding a creep-damage model for salt rocks.

Reading the manual, I saw that ABAQUS has a CREEP subroutine. In this subroutine, an creep deformation increment, DECRA(1), is defined and is multiplied by a vector n, i.e., the gradient of the deviatoric stress potential
(

https://abaqus-docs.mit.edu/2017/English/SIMACAESUBRefMap/simasub-c-creep.htm).

In my model, the DECRA(1) should be multiplied by the gradient of the damage stress. Because of that, I believe that the subroutine will not be correct.

Are there any subroutines in ABAQUS which the all strain components are defined?

 

[FEA way]

I’m afraid that in such case you will have to prepare UMAT (whole constitutive model).

 

[ttavio]

Yes. Do you have an example of a UMAT subroutine with creep-damage model?

 

[FEA way]

Check the work titled "Abaqus implementation of creep failure in polymer matrix composites with transverse isotropy" by F. Ouyang. It features whole UMAT code.

 

[Marekcom]

You need to implement an UMAT subroutine written in Fortran and linked to Abaqus Standard/Implicit numerical scheme.
Pay attention to the Jacobian matrix.

Structural Engineering

 

[ttavio]

Thanks Fea Way and Marekcon. I read F. Ouyang's work and decided, at first, to implement Norton's law for plane deformation and axisymmetric problems. In this case, where there is no damage, would the Jacobian remain the same as the elastic material?