Highlights
Accurate modeling of airbags deployment is becoming more and more important. In some cases, like Out Of Position crash tests, applying uniform pressure to the airbag tissue might not deliver the necessary accuracy. With FVM technology the airbag internal volume is discretized and variables like temperature and pressure are computed for each finite volume. Robustness and efficiency of FVM method are dramatically improved in this release.
Robustness is improved also for connectors and constraints:
| • | Tied contact with penalty formulation (Spotflag=26), is now compatible with eroded elements. |
| • | Spotwelds (LAW59) may have different tension/compression behavior (failure and yield stress) |
| • | Spotwelds with coupled tension/shear behavior (LAW83) have a new and more robust formulation for plasticity |
Several advanced material laws and failure criteria have been developed:
| • | Hyper-visco-elastic materials: |
| - | Stability improved for LAW42, LAW62, LAW82 and LAW69: |
| ▪ | Total large strain formulation |
| ▪ | Compatibility with Prony viscosity |
XFEM technology to model cracks propagation in windscreens has been enhanced:
| • | Multiple integration points thru each layer allow to better capture bending in each layer |
| • | Different material types per layer to better capture layer-dependent behaviors |
| • | Cracks paths directions and element deletion |
| • | Cracks visualization improved: triangular holes on display has been resolved |
Composites is another domain of major improvements; two new shell properties have been implemented: a multi-layer shell allowing different material types and multiple integration points per layer and a new composite shell with delamination prediction capabilities.
| • | FVM is computed at TTF only: saves cpu time |
| • | Vent-holes are identified with a specific name and output |
| • | Compatibility between /EREF, /REFSTA and /XREF; these 3 options are compatible as long as they don’t have any common nodes. |
|
| • | /PROP/TYPE51: multi-layer shell property allowing multiple material laws types per layer and multiple integration points per layer. |
| • | /PROP/TYPE17, plyxfem: new composite shell formulation based on Batoz shells using stack and ply input. Delamination can be calculated in each interplay, thanks to additional degrees of freedom are added for each interply to compute relative displacement of each ply. This formulation is proposed as a beta release. |
|
Spotwelds
| • | LAW59 tension/compression: yield and failure stress are different in tension and compression |
Foams and rubbers
| • | Extended stability for hyper-visco-elastic materials (LAW42, LAW62, LAW82 and LAW69): |
| - | Total large strain formulation compatibility |
| - | Prony viscosity compatibility |
Cracks propagation in windscreens
| • | XFEM and PID51 compatibility: multiple integration points per layer allows to model bending in each glass layer with higher accuracy. PID51 also offers the possibility to define different material types for each layer, all available with /FAIL/FLD, /FAIL/JOHNSON, and /FAIL/TBUTCHER. |
Failure models
| • | /FAIL/TAB1 has been enhanced with a feature to model instability due to diffused necking and with an option to model failure due to thinning. |
| • | Possibility to delete elements if failure occurs in middle or user-defined Integration Point |
|
| • | /MAT/LAW42: several issues corrected for shells |
| - | Incorrect time step calculation in starter |
| - | Possible instability at very high strains |
| - | Plastic strain results are wrong |
| - | Viscous part does not work |
| - | Very slow in comparison to LAW42, when Prony coefficients are used |
| - | Possible instability with shell elements when viscosity is taken into account |
| - | Stability of total strain formulation (Ismstr=10) for solids |
| - | Time step reported as 'Infinity' and 'NaN' in the starter output |
| - | Possible instability for shells |
| - | Possible non-physical elements deformation |
| - | LAW58 and /PROP/TYPE16 in //SUBMODEL gives Starter error |
| - | Function scaling not working in case of tabulated input |
| - | Influence of element system on results |
| - | Results reproducibility issue when elements are deactivated |
| • | /MAT/LAW70: incorrect results with Isolid=1, Ismstr=11, Iframe=2; affects v13.0 |
| • | /MAT/LAW27 and /PROP/TYPE11: shell stress tensor are empty |
| • | /MAT/LAW28, PID6, Isolid=17: possible negative internal energy |
| • | /MAT/LAW34: possible negative internal energy with cyclic loading |
| • | /MAT/LAW35: issue for unit conversion in /BEGIN card |
| • | /MAT/LAW2: strain rate output in animation files is null (affected versions: 13.0 and 14.0) |
| • | /PROP/STACK: joined model does not work |
| • | /FAIL/TENSSTRAIN: Engine fails when Istrain=0 |
| • | /FAIL/HASHIN shows strange behavior for some failure modes |
| • | /PROP/TYPE11: wrong output of thickness and layer position when Thick value is inconsistent with sum of the thicknesses |
| • | /TETRA4: Itet=3 may not work when using more than 1 SPMD Domain |
| • | /PROP/SH_ORTH: when stress output is requested for both upper and lower surfaces, no output is exported for upper layer. |
| • | /PROP/TYPE22 with elasto-plastic layers and elastic part material gives engine error |
|
