You are here
How to combine XFEM and UMAT in ABAQUS?
Wed, 2010-11-10 03:21 - Liu Jinxing
I have an idea: In abaqus, I want to simulate crack propagation by XFEM, and meanwhile I also want to define the stress update algorithm by writing a UMAT.
Anybody wants to share related experiences?
»
- Liu Jinxing's blog
- Log in or register to post comments
- 21756 reads
Comments
I think what you need is the UEL
You can use the UEL for XFEM with your own constitutive law.
Ying Li
Department of Mechanical Engineering
Northwestern University
Evanston, IL, 60201
Interested in your UEL idea
Thanks for your discussion, Ying Li.I have never used UEL in ABAQUS, even though I have often used UMAT/VUMAT. Therefore I have a few quesions as follows:
(1) Can UEL be used in both ABAQUS/explicit and ABAQUS/standard ?
(2) About the implementation of XFEM in UEL, I am still struggling with the additional degrees of freedom, can the inclusion of additional DOF be implemented in UEL?
(3) Is there any UEL learning material that you might share to me? It is alos best If there is an complete example
Don't spend time waiting for the perfect situation, something that is not very likely to come
thx for the answe do you
thx for the answe do you have an example as an input file
xfem abaqus
UPPALURI
http://imechanica.org/node/4338
UPPALURI
Since ABAQUS 6.9 has already an existing XFEM implementation....
Thanks for your reply, friends.
As we know, ABAQUS 6.9 or newer versions have implemented XFEM already. What I mean is to use the existing XFEM functions in ABAQUS, meanwhile to use UMAT to define specified material's properties. Can you talk about this instead?
About UMAT subroutine
Dear Liu,
I would like to ask you something about UMAT subroutine. I also want to couple UMAT with XFEM, but I am struggling how to transform my Euler angles to the 2 local vectors of UMAT. I am using a fortran code, but the results I get are quite different from the ones in literature. What equations are you using? And the second question is: are you using a code that changes the 2 vectors in every integration point?
Kind regards,
Panos
Euler angles
Dear Panos
If you are using the Euler angles to define the crystallographic orientation of a single crystal, there is a definition of a rotation matrix defined by the relationship between the global axis and the local(crystal) axis. But the rotation matrix has different forms according to the definition of the employed Euler angles(e.g., Bunge, Kocks, Roe etc.). If you need, I can send parts of my code via E-mail.
Sincerely yours,Sangyul Ha
About UEL and UMAT
Dear Ying,
I would like to ask you, instead of using UEL for crack initiation and propagation is it possible to apply a criterion within the UMAT?And another question based on your answer about UEL.
How is it possible to extract the required data from UMAT to UEL. For example let's say that I need a crack initiation based on accumulated shear strain. How can I related the corresponding value, with the element/node, the increment, and the neighbouring nodes?
Regards,
Panos
soryy about this question i
soryy about this question i am just a user of abaqus and need to use xfem
i am working with soil and my question how to produce a grack acording to compression stress not tenstion
and do we have the posibilty to use xfem with hypoplasticity constitutve low using umat.f subrotine
sory if my question is not good to the level of most of you but as i said i am just a normal user for abaqus
my best regards and wishes
sam
I had actually never heard
I had actually never heard about simulating crack propagation by XFEM performed in abaqus. I had previously heard about couple of rhspec done over the platform. But, had never heard about anything like this before. I would be really interested if you would go through the test and tell us briefly more about it.
XFEM and UMAT subroutine
I have UMAT soubroutine for ductile material damage model. When I try to link my subroutine with XFEM for predicting crack iniatiation and crack growth, the simulation is working, no error during running the analysis. But, unfortunately when I want to extract the result from solution dependent variable (SDV), it didn't give me a result. the graph plotted was totally wrong.
Have everybody have a same experience ??
Many thanks.