Hoping for a match.

Hello everyone,

Before starting simulating tensile tests on defective steel bars, I want to make sure that the software behaves properly : I define the behavior of my material by interpolating some points (true strain/true stress) of an experimental hardening function, then I run the software for a simple cylindric bar without defects.
The software should normally return a behavior curve similar to the one I submitted. Unfortunately, this is not the case. One can see the difference on the file attached (xlsx).
I took care to choose the best interpolated points that reproduce the experimental behavior curve as faithfully as possible, and I verified that my points are ok by plotting the interpolated hardening function on the software interface before running the computing. I also tried to change my points (choose more points at the end of the curve to insist on that part for instance) but the numerical behavior curve has always the same shape, the one shown on the attached file. I really don't understand why it does not behave properly through all the strain interval. May be it's the type of extrapolating that I selected (linear) ?

If you have any idea or comment, feel free to share it ;)

Regards.