[Paraview] "Cell Data to Point Data" with CFD wall values

[Magician]

Sam and Andy,


Thanks for your advices.
I found the order of applying filters is very important.
Here is a good step:
> 1. Extract Fluid block(s)
> 2. Cell Data to Point Data at Original Multiblock
> 3. Resample Data Sets of Original Multiblock at Fluid Block(s)

Most cases are OK with above step, but some cases including
non-contiguous interface failed on resampling step
(mismatching about Tuple Length).


Magician


On 2012/07/03, at 0:05, Andy Bauer wrote:

> What I would try is extract block filter to separate the 2 data sets. Then use the resample with data set to project the wall point data to the volumetric grid point data. I'm not sure if it will work without testing it though.
> 
> On Mon, Jul 2, 2012 at 10:39 AM, Magician <f_magician at mac.com> wrote:
> Hi all,
> 
> 
> I'm now trying to visualize CFD simulation data.
> There are 3D-Cell and 2D-Wall elements
> and each elements have cell-centered vector or scalar data.
> Wall and Cell elements are read as a single Unstructured
> Multiblock source.
> Wall elements are exactly overlapping with cell surfaces.
> 
> I often use "Cell Data to Point Data" filter.
> When I apply this filter for my Multiblock data,
> point values of 3D-Cells and 2D-Walls are treated separately.
> If possible, I want to replace point values of 3D-Cell surfaces'
> with 2D-Walls'.
> 
> Does anyone have a good idea to do it?
> 
> 
> Magician

On 2012/07/03, at 0:02, Samuel Key wrote:

> Magician,
> 
> If you were using EnSight Gold formatted binary results output, you could use the EnSight "Part" construct to output (both in the same results file-set) the 2D-Wall elements and the 3D-Cell elements as separate displayable graphic objects (plus their point & cell data).
> 
> Then, in PV, use the "Extract Block" filter to select which "Parts" you want to display followed by any other filters you like.
> 
> Sam
> 
> 
> On 7/2/2012 8:39 AM, Magician wrote:
>> Hi all,
>> 
>> 
>> I'm now trying to visualize CFD simulation data.
>> There are 3D-Cell and 2D-Wall elements
>> and each elements have cell-centered vector or scalar data.
>> Wall and Cell elements are read as a single Unstructured
>> Multiblock source.
>> Wall elements are exactly overlapping with cell surfaces.
>> 
>> I often use "Cell Data to Point Data" filter.
>> When I apply this filter for my Multiblock data,
>> point values of 3D-Cells and 2D-Walls are treated separately.
>> If possible, I want to replace point values of 3D-Cell surfaces'
>> with 2D-Walls'.
>> 
>> Does anyone have a good idea to do it?
>> 
>> 
>> Magician