Nonlinear behavior of steel materials

I have problem with 2 steel materials and finding theirs correlation with real tensile strength tests.

I've tested 3 materials (on tensile strength machine, according to: ISO 6892) and then I put material data into ANSYS and wanted to receive similar results.

To check when the sample "breaks" in ANSYS I use this Eurocode 3 equation "50*(Re/Young Modulus)" and compare it with plastic strain in ANSYS.

For material 1.4310 results are very good, real sample broke at 19.5kN and I got plastic strain about 0.2 at 19.9kN - that's correlation above 97% (but weird thing is that simulation crashes few substeps after this threshold - I think it shouldn't but maybe it's some clue to further investigation).

But for materials 1.4301 and 1.4509 I got worse results.

1.4301 in real test broke at 13.6kN and in ANSYS it "breaks" at 10,5kN - correlation: 76%.

1.4509 in real test broke at 9,5kN and in ANSYS it "breaks" at 8,1kN - correlation: 85%.


I created special excel file, to convert elongation and force (output data) into true stress and plastic strain for all 3 steels, sample:

Then for future analysis I took only True stresses above Re and only to maximum value (no decreasing) and I got this curve (sorry for no values but I can't give you them)

My testing setup:

I'm using two surfaces, one fixed and one allowed movement in Y axis - where I apllied force - all just to simulate grips in tensile machine.

On the other side I have displacement - where I also check applied force.


Does anyone had this kind of problem before?

Comments

  • jjdoylejjdoyle PittsburghMember

    It sounds like you are losing enough stiffness that the problem becomes structurally unstable.

    The nonconvergence might be indicative of the onset of necking.

    Is structure yielding thru entire section?

    Try plotting reaction force vs displacement. Does slope go flat or negative?

Sign In or Register to comment.