I have a general question: how do you define constantly changing material-coordinate-systems most conveniently in which commercial FE-software (or pre-processor)?
I know that IDEAS contains an option that enables the user to define the fibre-direction (the coordinate-system) along the surrounding surface of a yarn.
Abaqus for example seems to lack this option.
Any suggestions?
Material direction on curved yarns
Moderators: Martin, Developers
-
Wout
Hello anonymous,
Texgen can do the orientation lookup for a series of points. If you have access to the element centroid locations you could pass them through (the python or C++ function) and write a separate coordinate system for each element or for groups of elements with the same orientation (either using python interfaces or a separate prog operating on your FE input deck).
Alternatively you could have a "live" texgen model running that can do the querying on the fly (as in, pass the info from a User Subroutine and get the coordinate system back, then get on with the FE stiffness matrix assembly). However, this is a lot nicer when you have a C/C++ open FEM code than when you have a Fortran user subroutine in a commercial package.
Summarising, there is no point-and-click solution for that, and I am not sure if it will get there (integrated in Texgen), as it would require texgen to process a load of FEM input formats (which is hard to make fool-proof). Making a hack for your particular problem is probably done quickest with python.
Greetings
Wout
Texgen can do the orientation lookup for a series of points. If you have access to the element centroid locations you could pass them through (the python or C++ function) and write a separate coordinate system for each element or for groups of elements with the same orientation (either using python interfaces or a separate prog operating on your FE input deck).
Alternatively you could have a "live" texgen model running that can do the querying on the fly (as in, pass the info from a User Subroutine and get the coordinate system back, then get on with the FE stiffness matrix assembly). However, this is a lot nicer when you have a C/C++ open FEM code than when you have a Fortran user subroutine in a commercial package.
Summarising, there is no point-and-click solution for that, and I am not sure if it will get there (integrated in Texgen), as it would require texgen to process a load of FEM input formats (which is hard to make fool-proof). Making a hack for your particular problem is probably done quickest with python.
Greetings
Wout
-
Wout