I am not a NX user but I have a problem related to NX:
I have a custom post-processing tool which generates nodal scalar output for FEMAP in CSV format, which is basically:
node ID vec1 vec2 ... vecK
| | | |
| | | |
Is there a way to import this easily in NX? perhaps not coma-separated, maybe space separated?
If possible my goal is to avoid the UNV format, and stick with a format close to the one above.
I do not have acces to NX's help, sorry if the answer is obvious.
Thanks for your help,
To help determine how to proceed, can you tell us what you need to do with the results inside of NX. For example, do you need to generate a contour plot or an X-Y plot, or will you use the results as a boundary condition on a subsequent analysis, etc.
Thanks for the quick response!
I would like to generate contour plots of the scalar data.
I have no notion whatsoever of how NX organises output. The FEMAP CSV is organised with outputsets and outputvectors: the outputset is a collection of vectors, the vector is the actual output data
The only way to directly generate a contour plot from results is to use a .unv file. Since you've said that you want to avoid this, here are a couple of other options:
1) You can import CSV data into a Table Field. The display of a table field allows you to visualize the distribution and relative magnitude of the data:
However, there is no color bar and you cannot probe the values in this display. You can map a table field onto a boundary condition, such as nodal temperatures or element pressures. You can then contour plot the magnitude of that boundary condition as a standard Post Processing display that can be queried:
Hmm OK, I'll compromise and accept the UNV.
I'm only familiar with the dataset 2414 which was the custom tensorial output back when I used it.
It could have nodal and elemental output, but always written in 6 columns.
I would like to make it as similar as possible to the FEMAP CSV, espacially so I don't have to modify my programme's underlying architecture.
Here's what I have in mind, will this work? (I'm using the unv specification from Cincinnati SDRL, not sure if it's still applicable...)
Datasetclass = data at nodes
I have 64 output vectors, so for each node, 64 values
1) Can I write this in only 1 dataset 2414? (from the spec I assume yes - very important point!!)
2) Do I have to write it in 6 columns? (from the spec I assume yes)
3) How are titles handled? (as to that I have no clue)
Thank you very much for your help