[rsoni451]

Hello Everyone,

I ran a simulation in Converge CFD for 300 CAD to 750 CAD to analyse the combustion process for CNG direct injection in a single cylinder engine. I am using Ensight for post-processing and want to extract the value of different variables like u, v, w, temperature, pressure etc. at all CADs at the cylinder mid plane in text format for further processing in MATLAB. I know how to make a clip plane, but I am stuck here and really have no idea what to do next.

Any help will be highly appreciated !
Thanks in advance!

Best Regards
Rajat

[kevincolburn]

Once you have your clip plane created, make sure that your variables you desire to export are activated.

Then, just select the clip plane (part selection is pivotal in EnSight), and go to File -- > Export -- > Geometric Entities.

There are a couple of file format options, depending upon what you want to do. If the variables are nodal variables, you could select the "flatfile" format, specify a filename, turn OFF binary, and export. This would create a separate file for each timestep, which contains the nodes of the clip plane (the selected part(s)), which has the nodal positions (xyz) and the active nodal variables. If you want elemental data exported, make sure you have elemental variables, and use the parameter CELLID with the flatfile export.

The EnSight Case gold format is also open, and sometimes is useful, because it writes out element connectivity too.

With a transient model, using the File -- > Export -- > Geometric Entities, the time domain will automatically change, all of the clip will get re-calculated, variables re-computed, and exported.

-Kevin

[rsoni451]

Hello Kevin,

It worked! Thank you so much for the help!

Best Regards,
Rajat Soni

[kevincolburn]

I'd suggest selecting the parts, activating the variables of interest, making sure that you have Nodal variables (not Elemental Variables... see UserManual for ElemToNode if you do), and using the File -- > Export --> Geometric Entities -- > Flatfile format, binary option OFF, and write the data as flatfile. This will have x,y,z coordinates followed by variables. One node per line. This is not "matlab" format per say, but it is a file from which you can build an input into matlab to do further work.

Alternatively to this, you can calculate Tumble all within EnSight, if you want take that route.

-kevin

[rsoni451]

Hi Kevin,

I just came to know that I can directly plot the Tumble in Ensight itself by using In_cylinder_Tools (Python scripts).

It is written on the website that I need to paste the file in ".ensight100/extensions/user_defined/Tools/" directory but it seems like it belongs to the Windows version. I am using Ensight on Linux and not been able to find this directory. Can you please tell me where should I paste the in cylinder tools in a Linux system.

Kind Regards
Rajat

[kevincolburn]

On your linux system, you should have a directory in your home directory called ".ensight100", or ".ensight101", or ".ensight102", depending upon the version of EnSight that you are running. If you are running EnSight 10.1, you should have a ~/.ensight101/ directory.

[rsoni451]

Hi Kevin,

Inside ensight101 I am not able to find the "extensions" folder. Are you suggesting that I should create a directory "/ensight101/extensions/user_defined/Tools" ?

Best Regards
Rajat

[rsoni451]

I found it. It was inside site_preferences folder

[kevincolburn]

You can place them down in the installation directory ($CEI_HOME/ensightXXX/site_preferences/...).

This will effect all users who run using this installation.

For your own preferences, then you "should" have a preferences location set (this is typically ~/.ensightXXX/) on a linux system. However, you can verify this by starting up EnSight, and clicking on the Help -- > Version. In the dialog box which opens, there is a window at the bottom. About 9 lines into that window, it should say : "Preferences: /home/kevin/.ensight101/" This is the directory that you can step into, and you should find an "extensions" folder in there, and a "user_defined" folder under that one. Remember this is a "hidden" directory, as denoted by the preceeding dot "." in the directory name.

-kevin

[rsoni451]

Hi Kevin,

until now I have used MATLAB script to get the Tumble ratio by extracting column Data from post_convert. But, I want to use the in-cylinder tools for comparing the Tumble Ratio values.

My question is that the tool provided in the Ensight gives Tumble Ratio for the XZ intersecting plane along the Y-axis or for the whole cylinder?

If no, then what is the possible way to use this tool to get the mid plane Tumble Ratio ?

Bests,
Rajat

[kevincolburn]

The calculation is done in the parent parts. If you have the 3D volume for the cylinder as your parent part, then the calculation is for the 3D cylinder. If you have the 2D cross section as your parent part, then the calculation is for that 2D cross section. I've tried to stay true to the Parent Part paradigm in EnSight for this tool.
-kevin

[rsoni451]

Thanks Kevin!

I will give it a try.

Bests,
Rajat

[rsoni451]

Hi Kevin,

I procedure I followed to calculate the tumble on the mid plane is as follows.

1. Creating a clip plane.
2. Selecting the clip plane and then running the In cylinder tool.

the SpaMean_tumble that I am getting is in the range from -1.246 to 11.7527 which is obviously not correct as the value that I am getting from the matlab script is lying in the range of -0.5 to 3.5

Is there some thing that I am missing. Because both the curves look alike.

Kind Regards,
Rajat

[kevincolburn]

Rajat,
When you get the point of having a clip plane created, and the velocity variable activated, could you do the following for me:
a. select the clip plane.
b. File -- > Export -- > Geometric Entities -- > EnSight Case Gold format.
c. Specify a filename. (perhaps 'kevin')
d. to make the file transfer small, change the timestep range to the current timestep of interest (For timestep 45, specify 45 as begin, 45 as end, and 1 as increment).
Then ok.
This will write out ONLY the clip plane, and any activated variables to a small EnSight files (all named kevin*). Send me those files, and I can take a look at the routine with your data. In truth, you can also load JUST this dataset into EnSight, and run the utility. You will there be assured that the tumble calculation is done only on that part (as that is the only part in this small submodel).

-Kevin