FeFp Model in ABAQUS

[Jorgen]

The most recent extended functionality release of ABAQUS (v6.7-EF) contains a new model called the FeFp model. A lot of people have asked me about how this model works, and since the ABAQUS manuals do not provide much information I decided to performe a study of how this new model works for elastomers and thermoplastics. You can read the results from this study here, or in the attached document.

In summary, the FeFp model is claimed to be able to "predict permanent set in filled elastomers and thermoplastics". My study, however, shows that the FeFp model is not able to accurately predict the response of either elastomers or thermoplastics. In fact, the predictions from this model are not very different from the old metal plasticity models, which are known to not work well for polymers. I have also shown that much more accurate material models exist, for example the dynamic Bergstrom-Boyce model (DBB) for elastomers, and the three-network model (TNM) for thermoplastics.

[mbriere]

Thank you for evaluating this model. Wow....Brutal....but honest. I'd like to hear from the Simulia guys. What say you?

I know my company will be forgoing this model, at least for now....too much uncertainty.

Thank you Dr. Bergstrom.

[egal]

Thank Dr. Bergstrom for this terrific work.

My manager asks me to take a training course on this FeFp model. To be honesty, I doubt this model can model loading/unloading and permanent set so well. I would like to know how the constitutive equations look like.

Does anyone have any experience on this model? Please share with us.

[Jorgen]

I just received an official response from SIMULIA in regards to my review of the "Permanent Set Model", also called the "FeFp" model.

I have added this response to the pdf-file, and included it here as well:

The material model currently implemented in Abaqus is intended to capture permanent set in filled rubber subject to cyclic loading under multi-axial stress states and mild load reversal. Working closely with an industrial partner this capability has been validated against experimental results. Details of the validation procedure, including test data, input files, an Abaqus/CAE plug-in for calibration, and a paper presented at ECCMR 2007 are available through Answer 3522 on the SIMULIA Online Support System (www.simulia.com -> Support -> My Support). We acknowledge that the material model currently implemented in Abaqus is not suitable for complete load reversal and is limited in its ability to capture effects such as plasticity and rate dependence. We expect to relax these limitations and extend the polymer modeling capabilities of Abaqus in future releases.

-Jorgen

[vishalnayyar]

Hi Jorgen,

I saw the analysis you have done using FeFp model. I am interested in using this model but I could not find any documentation for this. Is it possible for you to help me regarding this.

Could you please send me any input file or any documentation about this model. Basically how to create the input file.

Thank you for your help.

Vishal
PhD student,
UT Austin
vishal.uta@gmail.com

[Jorgen]

Here is a template:

** FeFp model (hyperelastic + Mullins damage + Plasticity)
*Hyperelastic, Arruda-Boyce
** mu, lambdaM, D=2/kappa
%mu=1.0%, %lambdaM=3.0%, %D=0.002%
*Mullins effect
** r, m, beta
%r=2.0%, %m=1.0%, %beta=0.01%
*Plastic, hardening=isotropic
**(you can specify an arbitrary number of points here)
%sigY0=10%, 0.0
%sigY1=14%, %epsY1=0.01%
%sigY2=16%, %epsY2=0.04%
%sigY3=17%, %epsY3=0.20%

Note that the MCalibration software can calibrate that model (and many other models).

-Jorgen