Personal tools

Session 16

Behaviour model for semi-crystalline polymer, application to crashworthiness simulations
Today polymer materials are frequently used in the transport domain with more severe specification requirements. The behaviour modeling and failure prediction have consequently become a priority. In this paper, an elasto- viscoplastic behaviour model is presented, with non associated plasticity, damage and strain rate effect, which represents the observed behaviours of a semi-crystalline polymer under dynamic loading.
A constitutive equation for the aging of elastomer and application to dummy impact programs
A constitutive equation for chronorheologically simple materials that describes the aging and viscoelastic behaviors of elastomer is presented. A simulated numerical uniaxial relaxation test of a material at various aging stages has been performed. The simulated experimental results demonstrate the chronorheological effect and are used further to determine the material property functions in the constitutive equation. A test of an elastomer at various aging stages has been performed. It demonstrated the same effect as the simulated numerical example. The applications of this constitutive equation to dummy impact programs are mentioned.
A constitutive model for thermoplastics with some applications
A constitutive model for thermoplastics is outlined in this paper. The model consists of two parts: A hyperelastic-viscoplastic response due to intermolecular resistance denoted Part A, and an entropic hyperelastic response due to re-orientation of molecular chains called Part B. Both parts are developed within a framework for finite strains. The main constituents are the Neo-Hookean model describing large elastic deformations, the pressure-sensitive Raghava yield function, a non-associated visco-plastic flow potential and Anand’s stress- stretch relation representing the intramolecular stiffness. The 11 non-zero coefficients of the model are identified from uniaxial tension and compression tests on two materials, HDPE and PVC, which are respectively semi-crystalline and amorphous thermoplastics. Subsequently, it is employed in numerical simulations of three-point bending tests on the same materials. The model gives satisfactory predictions when compared to experimental behaviour.
Validation and Material Modelling of Plastics
The virtual estimation of physical product properties is only as good as the virtual description of the behaviour of its material. On the one hand there are well known material cards like *MAT_PIECEWISE_LINEAR_PLASTICITY in LS-DYNA© developed to describe a simplified behaviour of metallic materials. The reduced complexity of these material cards makes it possible to determine its parameters with less effort in actual material testing. Main advantages are high numerical stability and less machine time. On the other hand complex material models like *MAT-SAMP-1 can also handle varying compression and tension behaviours by defining a load case dependent yield surface as well as unloading by using damage functions. With the exception of visco-elasticity the description of visco-plasticity fulfills many requirements to describe a realistic behaviour of thermoplastics. For acceptable use of the above mentioned models a higher amount of load cases like tension, compression, shear have to be carried out to determine the material parameters and to represent the thermoplastic characteristics in crashworthiness simulations. At the moment there is no standardized method to determine material card properties for arbitrary material models from basic (i.e. tension, compression or shear) test setups. 4a impetus represents a standardized method, an efficient and reliable process starting with realistic test scenarios and finally ending up with a validated material card. The method of reverse engineering is used behind this process to generate material cards like *MAT_PIECEWISE_LINEAR_PLASTICITY as well as more complex *MAT_PLASTICITY_COMPRESSION_TENSION with regard to easy and favourable testing. We have compared different ways to determine and validate material cards with the example of PA6. Limits and opportunities of different test methods and material card implementations are shown and compared to each other especially focused on typical polymer behaviour.
A new advanced visco-elastoplastic eight chain rubber model for LS-DYNA
A new advanced eight chain rubber model has recently been implemented in LS-DYNA. The material is tailored for polymeric materials. The basic theory is taken from Arruda’s thesis from 1993 but it has been enhanced with advanced features such as the Mullins effect, viscoelasticity, plasticity and viscoplasticity. The Mullins effect is described by two different models: the first one is strain based and developed by Boyce in 2004 and the second is energy based and developed by Ogden and Roxburgh in 1999. The viscoelasticity is based on the general Maxwell theory with up to six Maxwell elements (a spring and a dashpot in series). There are three different viscoplasticity models implemented: a Norton model with two parameters, a G’Sell model with six parameters and a strain hardening model with four parameters. The plastic yield strength is based on the eight parameter Hill model. The material model has been used to simulate a compression test with a rubber specimen. The material parameters were obtained from inverse FE analys and parameter fitting using LS-OPT and a force-displacement data set. The result shows that this material model can predict rubber behaviour inline with experimental results.