Getting Force-Displacement curve in the contact examples

[bmurgas-LANL]

Hi Moose community,

I am new using the contact capabilities of Moose. I just started to run the examples and tutorials available in the Contact module.
In the following examples there is a Force-Displacement curve that I would like to obtain:

https://mooseframework.inl.gov/modules/contact/TwoDimensionalSphericalIndenterMortar.html
https://mooseframework.inl.gov/modules/contact/TwoDimensionalSphericalIndenterNodeFace.html
https://mooseframework.inl.gov/modules/contact/BerkovichIndenterNodeFace.html

Could you explain how to get this plot? Please.

Thank you very much.

Best,
Brayan

[GiudGiud]

@recuero could you please share that plotting script if you still have it?

[recuero]

This was obtained directly from MOOSE's postprocessors, using the resulting reaction force on the prescribed Dirichlet BC on the indenter and the imposed displacement (which should be an input to the model). As far as I recall, that was everything needed.

[bmurgas-LANL]

Hi @recuero,
Thank you for your help.
Yes, I see that in the input files you have this postprocessor to get the displacement:

[Postprocessors]
  [maxdisp]
    type = NodalVariableValue
    nodeid = 39
    variable = disp_y
  []
  [resid_y]
    type = NodalSum
    variable = saved_y
    boundary = 1
  []
[]

But there is not a postprocessor to get the force. I tried to compute a force using the average stress on the indenter and dividing by the surface and other postprocessors but I don't get the correct force. Could you indicate me which postprocessor was used? Please.
Thank you!

Brayan

[recuero]

That saved_y variable used to (maybe it still does) provide the overall reaction force. If it changed, there will be a new method to do so.

[bmurgas-LANL]

Thank you for your help.
I got the results for the examples using the node-on-face formulation:

I got confused because I wasn't sure I could get the force from the residual but I finished reading the paper of the formulation and now is clear.
Also, the dimensions of the plot were changed to [kg m s-1] and we get [kg mm s-1] from the simulations which was kind of confusing at the beginning and the displacement from the initial contact was subtracted for the rest of the positions:

Thanks again! :)