Hello everybody,
My task is to simulate non-linear viscoelasticity. For now, I have written a UMAT based on Schapery's model. To verify it, I am trying to perform relaxation test in Abaqus. These are the steps that I follow
I use a single element mesh on the 1D Truss of element type (T2D2). Attached is my input file.
I would really like to hear from people and please correct me if I am wrong.
Thanks in advance for your time
Shree
Hallo Shree,
Relaxation test is one of the most complicated tasks, Many information must be gathered in order to model such a case. what you have done is pure static with small deflection. the following are important factor in performing relaxtion test:
-aim is finding viscoelastic properties (Prony series parameters)?
-non-linear material model?
-experiment time?
-Stress variation regarding boundary condition?
Good luck
Hello Mr. Qaimari
Thanks for the answer. With respect to your questions,
- I am not performing the relaxation test to get Prony series parameters. In fact, I have written a UMAT based on a paper (attached). In the paper they have Prony series parameters. They also have sample relaxation graphs for 3D elements. To start off, I coded the 1D formulation of the constitutive model. The relaxation test was just to make sure I am able to simulate the general behaviour in 1D so that I can extend my code to 2D cases.
- Btw, is it necessary to use Visco steps or Static,General works as well?
Looking forward to your answers
Thanks and Regards
Shreeraman S
Hello Shree,
I am having a problem with the Creep experiment test perform in Abaqus environment. I am using Chaboche model. For that i already have UMAT for it. i just want your help in abaqus. How to perform my the creep test in abaqus to validate my experimental test.
Thanks and regards,
Anil Patel.
HI Anil,
I have not performed a creep test in Abaqus. I somehow prefer (disp based) relaxation test to creep and they provide the same (similar) information in terms of material behaviour. For creep,
- In case you are not in large scale deformations, I would suggest to apply a heavi-side force based step load.
- That should in principle give you a constant state of stress after the peak stress which you can post-process and verify.
I shall start that way and see how it goes. I could not think of anything better right now. Sorry!
Anyway good luck and regards
- Badges
- Science topic