Non linear material model not working as expected

Cat69

Hi,

I used the commercial version of Mecway in a previous workplace and I have the free version on my home computer.
I did a simple cantilever beam test study with material as elastic-perfectly plastic (Yield stress0=200 MPa). Material was modelled as plastic "Isotropic hardening curve" and the study is non-linear quasi static, with load increasing linearly from 0 to max value.
The maximum resulting stress should be limited to the Yield stress (200 MPa), however it shows values above Yield, and it behaves elastically.
Could anyone tell me what I do wrong or is it program's limitation?
File attached.

MikeMcMullen

I think it is a limitation of the implicit solution process in general as often there is a problem with converging on a solution if the resistance does not increase at least a little with increasing strain. Therefore the material models are not perfectly plastic. Also a problem when the resistance decreases with increasing strain. Explicit methods can work with these materials and are often used for dynamic non-linear problems. In these problems inertia and sufficiently small time steps provides the increasing resistance needed for convergance.

disla

If you look at the beam integration points you will find that VM Stress is below the yield point. Node values are extrapolated and are not accurate enought for the element/mesh selection.

stresses (elem, integ.pnt.,sxx,syy,szz,sxy,sxz,syz) for set ESET1 and time 0.1000000E+01

sxx syy szz sxy sxz syz VM
1 -171.01 -0.12 -1.44 -0.27 0.08 -0.16 170.23
2 -171.06 -0.17 -1.55 -0.09 -0.33 0.15 170.20
3 -171.01 -0.12 -1.44 -0.27 -0.08 0.16 170.23
4 -171.06 -0.17 -1.55 -0.09 0.33 -0.15 170.20
5 171.18 -0.17 1.45 -0.73 0.33 -0.15 170.55
6 171.23 -0.12 1.55 -0.91 -0.08 0.16 170.52
7 171.18 -0.17 1.45 -0.73 -0.33 0.15 170.55
8 171.23 -0.12 1.55 -0.91 0.08 -0.16 170.52

VM Max 170.6 MPa

Cat69

Thanks, disla and Mike for your comments.

The calculated theoretical elastic bending stress is 300 MPA, above Yield=200 MPa.
The program calculated von Mises and directional xx stress 285 MPa.

I wonder if anyone has encountered this issue or could share a simple example of elastic-perfectly plastic material, subject to bending.

I did the same example, but the beam was subject to tension, instead of bending and it behaved as expected, model became unstable above Yield stress.

disla

ccx Manual says classical von Mises-Huber yield condition applies.
This condition is evaluated at the integration points.
s.xx are also below 200MPa there.

Plasticity is better captured with solid elements.

https://calculix.discourse.group/t/elasto-plastic-bending-elements-accuracy/700/3