|
[Sponsors] |
CFD-Post: Polyline Export at Integration Points |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
July 1, 2011, 09:49 |
CFD-Post: Polyline Export at Integration Points
|
#1 |
Member
Join Date: Apr 2009
Posts: 36
Rep Power: 17 |
I am trying to write a clever script which will conveniently show me my boundary layer resolution among other things. So far, I have successfully been able to get polyline files containing normal lines to my surfaces. The issue I'm facing is that I only want data at the integration points.
Currently, I have to hand define how many points the polyline contains. CFX will then give me data at each point that I define. So, when I plot, I can "see" the integration points by where my line bends, but I have data points in between that looks linear interpolated. From an ease of scripting/charting, I would like to, e.g. define my polyline by only two points. Then, have CFX give me interpolated data at all the faces it crosses. I know this is. . .well quasi-possible, because Post does this when you do a polyline by boundary intersection. If you do this, you clearly get only the integration points in your polyline. Due to geometry, it's simply not possible to create my polylines in this manner. If anyone else has another suggestion as to the best way to get this, I would greatly appreciate it. If worse comes to worse, I can write a script to calculate the derivatives of my lines, and where the derivative changes, I will assume that is an integration point, and data reduce in that fashion, but I'd just rather not. Thanks, |
|
February 14, 2014, 09:56 |
|
#2 |
Member
Daniel Ceglarski
Join Date: Sep 2012
Location: Essen, Germany
Posts: 50
Rep Power: 14 |
Hi minger,
did you solved the problem? I am currently having the same Problem. Greetings Daniel |
|
February 15, 2014, 19:14 |
|
#3 |
Member
Join Date: Apr 2009
Posts: 36
Rep Power: 17 |
I think "a" way that you can do it is instead of defining the line as "sample" you define it the other way (not sure what the option is, but the line will span your domain). Then you have to isoclip the line. The isoclip will only show results where you want.
|
|
February 17, 2014, 08:16 |
|
#4 |
Member
Daniel Ceglarski
Join Date: Sep 2012
Location: Essen, Germany
Posts: 50
Rep Power: 14 |
Okay when I use a line object instead of a polyline, then I can select the "cut" option. This will expand the line over the whole domain. The Iso Clip method will now limit the portion of the line. Okay but the Ansys doc says:
>>Iso clip locators interpolate values using a method that is slightly less accurate than that used for slice planes and isosurfaces.<< So I better stay with Polylines and let CFX-Post interpolate the data from it. But thank you minger! |
|
February 22, 2014, 10:23 |
It works!
|
#5 |
Member
Daniel Ceglarski
Join Date: Sep 2012
Location: Essen, Germany
Posts: 50
Rep Power: 14 |
Just for information, the method stated above works well. The interpolated values on the polyline indeed seem to be linear interpolated (pls. have a look at the blue dots in the attached picture). If these additional interpolated values are not needed, one can go with a line instead of a polyline and just define two points that run along the line and clip it with Insert/Location/Iso Clip to get only data Points where the line intersects with a mesh element face (red circles in the picture). There seems to be no loss in accuracy.
Thank you again minger for the hint! |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
How to visualize Output Equation Residuals in CFD Post | rubin4ik | CFX | 2 | June 22, 2011 07:37 |
Creating Face in CFD Post and export Values as Boundary Conditions | flow87 | CFX | 2 | June 4, 2011 05:06 |
OpenFOAM on MinGW crosscompiler hosted on Linux | allenzhao | OpenFOAM Installation | 127 | January 30, 2009 20:08 |
Integration Points and normal area vectors | Bloshchitsyn Vladimir | CFX | 0 | November 26, 2007 08:35 |
Can you post a blade data points file example??? | Nicola | Fidelity CFD | 0 | September 21, 2007 12:38 |