[vtkusers] Best way to handle Gauss points?

[John Platt]

Hi,

Typically, FE solvers sample the stresses at optimal sampling points (which 
may coincide with the Gauss points) then linearly extrapolate to the nodes. 
If it is appropriate to average the stresses at the nodes, I would do this 
in your solver and write point tensor data to VTK. If the stresses cannot be 
averaged, you can create multiple VTK points at your nodes (with the same 
coordinates) for each element and set the topology accordingly. Point tensor 
data can then be written. Be careful that VTK does not remove coincident 
points.

HTH

John.


----- Original Message ----- 
From: "Mohamed Moussa" <mohamed.moussa89 at gmail.com>
To: <vtkusers at vtk.org>
Sent: Tuesday, September 09, 2014 5:02 AM
Subject: [vtkusers] Best way to handle Gauss points?


> Hello,
>
> I'm trying to output FEA results from our solver to a VTK unstructured
> grid format so we can visualize in Paraview. I'm stuck on stress
> tensor output. I'm writing C++ code and I've figured out how to output
> cells/elements, displacement results at the node, etc.
>
> Whats the best way of handling tensor data defined at Gauss points so
> that I can visualize interpolated results in Paraview in a
> straightforward manner?
>
> I have read that VTK quadrature document but I don't want to go down
> that path if it means I will have to write a filter in Paraview anyway
> (I noticed there is no 'interpolate/extrapolate from quadrature
> points' filter). I might as well just do it within my own program and
> output nodal data.
>
> Mo
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at 
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the VTK FAQ at: 
> http://www.vtk.org/Wiki/VTK_FAQ
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/vtkusers
>