Why is there an offset in the curve for input data and history output? [Abaqus CDPM]

[drennon236]

I am simulating a single element uniaxial tensile and compressive stress test. Then I compare the stress strain curve from abaqus, and the raw input data for inelastic compressive and tensile behavior. So the material data used in abaqus is the same plotted in the blue curve.

Test setup, the four corners are elongated or compressed using boundary conditions:

These are the results: Orange is the stress strain curve from abaqus.

There seems to be an offset in the peak stress for both compression and tension; does anyone know why?

 

[FEA way]

Try with the same symmetry boundary conditions as those used in Verification Guide for CDP model.

 

[drennon236]

Hey FEA way, this is using the verification guide model. I just copied the inp file. Do you think this has to do with the boundary conditions?

 

[FEA way]

Then you may have used files from different verification guide example. Here's how the BCs look like in case of the one I'm talking about:

 

[drennon236]

Hey FEAway, I found my mistake, it was because I started my blue curve from the inelastic strain, which starts at 0, but should start from the yield strain.