creep modelling of plastics using LS-DYNA

[bates]

I have been using MAT 24 to model the behaviour of polyproylene and polyethylene. The results have been reasonable, however I need to be able to give details on deflections over a period of time and not just instantaneously. I need to be able to predict ductile failure (i.e. to provide a safety factor against ductile behaviour).

Three material models available in LS-DYNA which appear to be suitable for creep. The first (MAT_VISCOELASTIC) seems simple, the second is more a general model MAT_GENERAL_VISCOELASTIC and the third, MAT_PIECEWISE_LINEAR_PLASTICITY is the material model I currently use for Instantaneous loading regimes, this model has an option for VISCOPLASTICITY and of the three may be the most suitable.

I would appreciate your guidance on the following questions:
(1) In your judgment, do any of these models seem appropriate for a general range of loading regimes (based on temp, duration and magnitude) ?
(2) For the best model available, how easy would it be to generate the input information ? what tests do I need to carry out?
(3) Are there any user developed subroutines available which may be suitable?

Thanks in advance,

Stuart

[Jorgen]

Hello Stuart,

As you know, PP and PE are important commercial materials that exhibit strongly non-linear behavior when exposed to thermomechanical loading. Another complicating factor is that it is in general more complicated to predict failure of these thermoplastics compared to traditional metals.

The usefulness (i.e. accuracy) of the three models that you mention will depend strongly on the magnitude of the applied strains, the nature of the loading, and the temperatures that you are interested in. None of the three models will be a applicable for general simulations - you will always need to be cautious with the material parameter calibration and validate your results. These models, when implemented and calibrated correctly, can give quite nice results. There is always a danger to run into problems, however, when using material models that were specifically developed for thermoplastics.

It should not be too difficult to calibrate any of those models. I would use a combination of uniaxial tension and uniaxial compression, at different strain rates. In addition to these tests I would also perform a few stress relaxation or creep tests.

I have developed a few user material subroutines that provide significant advantages in terms of accuracy and general applicability. Depending on your needs, those models might be of interest to you. For more information you can check out the User Subroutines (http://www.polymerfem.com/modules.php?name=User_Subroutines) section.

- Jorgen