[Paraview] transient data

[SamuelKey]

Oliver--

I have attached an example of vtk xml-formatted results that I generated 
from a transient dynamic simulation of two pipes impacting at right 
angles to a separate e-mail (The Kitware PV mailing list will reject 
this reply when this example data set is attached).

One pipe is "steel," and the other pipe is "aluminum." (The contact 
algorithm was supplied with a pipe wall-thickness and, therefore, they 
appear to not quite touch.) This example is only ten time steps. (I 
usually generate between 100 and 200 time steps.)


Short Instruction --

In ParaView, start by opening the *.pvd file. At the bottom of the 
display panel click on the ">" button of the "VHS player."


Long Instruction --

When I generate simulation results I separate the model into parts by 
material. Each "part" (.m00x.) mimics a piece of a parallel data set.

Each "time-step" (.t00x.) is in a separate file. This is a requirement 
of the vtk data reader.  And the answer to your next question is "Yes, 
the model connectivity and coordinates are repeated for every time step. 
You can write current deformed object coordinates to save yourself the 
trouble of invoking the Warp filter. The Warp filter can still be used 
to scale-up further the deformation."

All of these files are referenced by a single *.pvd meta-file. (pvd = 
ParaView Data)  The *.pvd file is an xml-formatted ASCII-text file; you 
can look at it with a text editor. The name="abc" token is as yet 
undocumented; it shows up on the Extract Datasets filter's face sheet. 
The name="abc" feature is very helpful if you have a large complex model 
and wish to isolate one-or-more parts for close inspection. Otherwise 
the face sheet only identifies the parts by number.

This file-set is dos-formatted; the file-set was generated on a WinAMD 
box. I have tested this file-set with ParaView version 2.4.4 and 2.6.1. 
For a Linux box you may need to convert the file-set with the utility 
dos2unix.

Lastly, my current preference is for EnSight-binary-formatted results 
which ParaView reads. They are much more compact, and each point-date 
item and each cell-data item are placed in separate files with all time 
steps.

I am happy to e-mail to you both or either Fortran-90 subroutines that I 
use to generate results for ParaView to display.

Enjoy,

Sam



Oliver Gloth wrote:
>
> Hello all,
>
>  
>
> I have been using ParaView for quite some time no to visualise CFD 
> data. So far I have created multiple data files to visualise transient 
> data:
>
>  
>
> Foo_001.vts
>
> Foo_002.vts
>
> Foo_003.vts
>
>  
>
> ParaView used to automatically enable the filename as a parameter for 
> an animation. Apparently this feature is not supported in the current 
> release, but will be back in the future. Is there a better way to deal 
> with transient data files? What would be the recommended best practice 
> to create an animation from a transient simulation?
>
>  
>
> Thanks for any advice,
>
> Oliver
>
>  
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> ParaView mailing list
> ParaView at paraview.org
> http://www.paraview.org/mailman/listinfo/paraview
>