ABAQUS VUMAT for shell element

[CF]

Hi

I'm coding a hyperelastic vumat for a shell element. I'm testing it by comparing simple one element uniaxial/biaxial results for the shell element vumat and a solid continuum element.

Everything checks out fine except for the out-of-plane strains (call it LEop). Whatever the dimensions/thickness of the elements, LEop for the shell element case is always twice the value for the continuum element case.

The strain increment in the shell vumat is calculated by:

strainInc(i,3) = stretchNew(i,3) - stretchOld(i,3)

I believe this is how ABAQUS calculates it for the solid continuum element case (debug print statements in the vumat confirm this).

Can anyone point me in the right direction?

Thanks
Cormac

[hhspiny]

have you check the convertion between tensor shear strain and engineering shear strain? i.e. engineering shear strain = 2 * tensor shear strain. abaqus uses tensor strain in calculation but reports engineering strain.

[CF]

Thanks for the response. Shear strains are all zero for my case.

[Jorgen]

Two comments:

:arrow: Can you calculate what the out-of-plane strain chould be? Can you verify that that your model implementation is correct based on this?

:arrow: For shell elements you need to explicitly calculate the out-of-plane strain that gives zero out-of-plane stress. The strainInc(i,3) variable is passed in for information, but it is up to you to actually calculate the out-of-plane strain.

- Jorgen