CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Post-Processing Results for Velocity Profiles (https://www.cfd-online.com/Forums/cfx/108542-post-processing-results-velocity-profiles.html)

Crank-Shaft October 25, 2012 22:21

Post-Processing Results for Velocity Profiles
 
Hello Everyone,

This is my first post here so I will try my best to describe the problem. I should be able to post images shortly.

Domain - 3D 16 Deg. rearward facing ramp with a total domain length of 2.5 m. Assigned inlet Velocity U_0 = 4. m/s and Reynolds Number of 81 000. The outlet pressure is 0 (ambient) and the top wall is modelled as a zero-shear wall. Both sides are set to symmetry.

The aim is to induce some flow separation and hence downstream Wall-normal lines have been created at 50 mm, 100 mm and 200 mm. I would like to plot Vel_U at the 3 locations on a single chart. The Y-axis needs to be the Y-coordinate (range 0-233mm) however, I am trying to present the X-axis with the 3 velocity profiles at the 3 locations respectively.

I understand that the units might be very confusing for this however, I think it has been done in literature with TecPlot. Please comment on whether this is possible using ANSYS CFD Post and I will really appreciate the help.

ghorrocks October 26, 2012 06:29

This is easy in CFD-Post. Define your points for the lines to be perpendicular to the walls and tiy can extract the velocity along the lines. I think that is what you are looking for.

Crank-Shaft October 26, 2012 21:57

2 Attachment(s)
Thanks for the reply Ghorrocks however, I don't think it is quite that simple. I have already drawn normal lines lying on the central, longitudinal plane of the ramp and the velocity profiles are easy to extract and export as CSV files.

I am trying to overlay these velocity profiles at the respective locations downstream such that the development of the flow is clearly and easily visualised. I have attached and image of the ramp geometry with the flow in (-x) direction [Capture.png]. The Profiles have been extracted and attached as an image [Capture2.png].

Is there a good way to create this through the user-interface or do we have to create some new expressions and variables? I have thought about making some expressions like x+u_{x} or something similar however, that is non-sense and the units would obviously be incorrect.


Thanks and appreciate the time.

ghorrocks October 27, 2012 07:28

In that case I do not understand what you are trying to do. Can you post an image of where you want to place the velocity profiles and what you want to do with it? Also I do not know what you mean by x+u_{x} - what is the purpose of this?

Crank-Shaft October 28, 2012 02:46

1 Attachment(s)
I have found an example of flow through a planar asymmetric diffuser.
The attached image was taken from a study conducted by Reid A. Berdanier of Purdue University. (April, 2011)

The idea is to present the data for velocity profiles at the different downstream locations along the diffuser.

Please advise further if you think this is possible within ANSYS CFD-Post.

Regards,
Ovi

ghorrocks October 28, 2012 05:57

I understand now.

I would be looking at tecplot or similar for doing something like this. This plot would be tricky in CFD-Post, but you could probably do something similar.

Crank-Shaft October 28, 2012 19:04

Yeah I think the result I posted previously was processed using TecPlot. I currently do not have access to this software.

Can someone please provide some alternative post-processing applications which can produce similar output? I would prefer an open-source or free application that accepts CFX or Fluent Data.

Thanks for any comments and suggestions.

ehsan210 September 18, 2016 03:24

3-D axial velocity profiles in ANSYS CFD post
 
1 Attachment(s)
Greetings Dear Friends
I want 3-D axial velocity profiles from ANSYS CFD Post down. So I can not compare my results with the photo attached.
Is there such a possibility in ANSYS CFD Post?

ghorrocks September 18, 2016 07:39

Try a user surface with an offset from an existing surface. Start with the cross section plane and offset by the velocity.


All times are GMT -4. The time now is 21:01.