There are many different models for predicting creep behavior in Ansys or Abaqus software. For example, one simple model is Norton-Bailey's model: [creep_strain=C(Stress)^m.(time)^n], which C,m and n are the creep constants. So, my question is that, Can we obtain the creep constants with the data of the tensile creep tests, if we want to simulate or model the creep behavior of a general structure which is on tensile, shear or compression loads?
Yes, you can derive the model constants by reproducing the data produced by tensile tests. You should make sure that the behaviour remains isotropic throughout, otherwise the fitting for other stress-strain paths may not be as good.
Dear Dr. Joaquin Marti
I already performed some tensile and shear creep test on a polymer, but as I want to simulate the shear creep in FE software with that tensile behavior, the results comes more than 4 times up than the real experimental shear test. and that's why that I am in doubt whether I shouldn't use the tensile creep model to predict the shear creep.
Hi Ata Khabazaghdam,
Are you sure, your material remains isotropic in creep loading.
In the case of metals, tensile creep models work for shear creep.
Dear Manish Kumar, I performed my tensile and shear tests in two different testing machines. I am not sure about the calibration of that both machines. after checking that, maybe I can decide that this phenomenon relates to the different properties of polymer in tensile and shear or maybe the size effect of dogbone samples.
Ata,
Short of calibration errors, the inconsistency of results points to a difference in behaviour. Size effects could be part of it, but I suspect anisotropy is more likely the reason, either preexisting or developed in the course of the creep process.
Hi Ata,
I am also agree with Marti's response, Anisotropy may be the reason of these results. There are good chance of pre-existing anisotropy in the case of polymers.
In my view, Elastic moduli are critical point, not only Anisotropy but also the differences between instantaneous modulus and long-term-modulus.
Wengang,
If the time evolution were the same for all moduli, tensile creep results would serve to predict the shear creep behaviour. For a mismatch to occur, you need anisotropy, i.e. differences in the varios moduli, either from the onset (pre-existing) or developed in the tests (induced during creep).
- Badges
- Science topic