[Gmsh] Stress on Gauss Points

[Max Orok]

Hi Jeremy,

My impression going in was that the Gauss point steps were part of the
simulation and the goal was just to plot the data.
I think for this data set, the shared node data does repeat, but a single
element will vary as a function of its nodes.

For any averaging etc., perhaps this could be done during the simulation?

Sincerely,
Max


On Fri, Feb 15, 2019 at 5:19 AM Jeremy Theler <jeremy at seamplex.com> wrote:

> On Thu, 2019-02-14 at 10:56 -0500, Max Orok wrote:
>
>
> Jeremy,
>
> You're right, there isn't any explicit averaging going on, it just goes
> through all the nodes and assigns values based on their number.
> After assigning I trusted Gmsh to display them nicely :):)
>
>
> ok, I now finally understood your script... I made up a simplified 2d
> version, see attached
>
>
>
> This does mean that node data values are set multiple times -> to the same
> value (magnitude at least).
>
>
> yes, I can see... nodes 2 and 3 get scalars 5 and 8, whilst I would expect
> and averaging between 2 and 5 in node 2 and 8 and 3 in node 3 :-/
>
> A lot of the data on each row is duplicated elsewhere in the file, my idea
> was that if the elements were ordered correctly, the
> repeated overwrite wouldn't matter as long as the overwrite contained the
> same number.
>
>
> this is the point, the data is not repeated! stresses or whatever
> evaluated at the gauss points in different elements will surely contain
> different values, even if they are near the same node... am I clear?
>
> I am interesting in seeing how to average these values to assign them to
> the shared node...
> plain average? weighted average? how? using the element's volume? the
> element's quality? etc
>
>
> regards
> --
> jeremy theler
> www.seamplex.com
>
>

-- 
Max Orok
Contractor
www.mevex.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://onelab.info/pipermail/gmsh/attachments/20190215/43fb9ad5/attachment.html>