ABAQUS Shell Local Coordinates

[johnbn]

Hello all,

I've been using 3D continuum shell element (SC8R) to model elastic deformation in a global sphere. I'd like to express the stress directions in global coordinates rather than in the local shell coordinates abaqus provides (S11, S22, S12; S33 is radial and is always set to zero).

I accessed what abaqus terms the local coordinate system values in the odb file through the fieldOutput variable via python commands. In the scripting manual it says the variable localCoordSystem is a matrix of direction cosines relating the local coordinate system to the global coordinate system.

If this is the case, then using this rotation matrix (R) of direction cosines I should be able to get the stress values in terms of global coordinates by the following procedure:

Stress_global = R*stress_local*R'

However, when I do this the global stress values I get do not make sense at all.

Is there something I'm missing here or should the procedure I'm using give the correct stress values in terms of global x-y-z coordinates?

If someone knows a better way to go from shell local coordinates to global xyz coordinates any suggestions will be greatly appreciated!!!

Many Thanks,
John

 

[TGS4]

I've learned this the hard way (and by talking with the ABAQUS help folks). Prior to performing the analysis, you need to create a material orientation that is aligned to the global coordinate system (although this may prove to be difficult because the local material orientation will try to force one of the coordinate directions to be normal to the shell).

Then, in the output, the "default" material orientation will be the one that you defined. Good luck.