Increase in plastic strains after unloading

[Gov45]

Hello all

I have a case in ABAQUS calculation where I applied the load on a bracket (bracket is attached to a frame, and bracket and frame are welded together), there is plastic deformation after loading on the frame. After unloading, the plastic strains are increasing on the frame. But as per theory plastic strains after unloading should be same as plastic strains after loading.
Elements used in ABAQUS --> shell elements (S4R)

loading --> max plastic strain 0.009
unloading --> mas=x plastic strains 0.014

Does this have to do anything with the elements?

Regards

 

[FEA way]

What settings are you using for plasticity (model, type of hardening, additional effects and so on) ? How are you reading the strains - from the contour plot or directly from the integration points ?

 

[Gov45]

Plasticity model - von Mises Plasticity
Hardening - default isotropic hardening
no additional effects are provided
reading strains from contour plot

 

[Gov45]

If anyone has come across such a scenario, please do reply to this thread.

 

[SWComposites]

would help to see:
- picture of the parts/assembly
- picture showing the loading
- plot of the stress-strain response at the peak location
- strain contour plots at various loads

first thing that comes to mind is differential yielding between different locations resulting in increased local strain at some yielded locations during overall unloading

 

[FEA way]

What if you disable averaging / enable quilt plot and check values at the integration points with the Probe values tool ?

 

[ThomasH]

If I speculate a bit, could be completely wrong .

When the load is applied you have a total strain that consists of an elastic part and a plastic part. When you unload the elastic part should disappear but could not that change result in an increased plastic strain at another location. I guess is that in may be due to the geometry of the bracket, or how the material model works. Or I could be wrong.