Profile Boundary Condition

sufjanst · August 7, 2017, 10:20

Hello,

I try to simulate a combustion problem.
I already simulated a domain with 4 burners. Now I'm trying to simulate a domain with 8 burners which are the same as in the first domain but on other positions. I created a Hexa-Mesh to have a complete symmetric domain. I wrote a profile for one burner and tried to use it as a boco for each of the 8 burners. This didn't work. Is that even possible due to the fact that the mesh positions are different at the 8 burners? Is there a solution for this problem?

Here are the positions for the 8 burners.

https://picload.org/view/rpppapgw/2.png.html

Sufjanst

sbaffini · August 7, 2017, 10:28

If by profile you mean a profile file for the inlet, than I'm more than 99% sure that it is at least one of your problems. Those files are meant to be used as interpolation tables, which depend on the coordinates of course.

Your best option would be to manually rotate your original domain and produce a new file for the new position, which will superpose with the rest of the final domain, then use both profile files.

Otherwise, if you are into it, you can write a small script (say, Octave/MATLAB) that will read the file, rotate the coordinates, and write it back. I don't know if recent Fluent editions have a feature for directly doing this on the files... I guess not.

sufjanst · August 14, 2017, 07:10

thank you for your reply. I wrote a Matlab script, which rotated my coordinates and profiles. I used them as inlets and they work!

I now have a problem that a few cells have very high velocities. All of the inlets have a velocity magnitude of about 20 m/s. Except for 10 cells which have a velocity magnitude of 1044514 m/s. I don't know why. I tried every interpolation method...

sbaffini · August 15, 2017, 16:13

How did you check that your script to rotate the coordinates was correct? Looks like it might have a singularity for some coordinates. Can you post it here?

sufjanst · August 16, 2017, 16:59

The coordinates are correct. I created a line with two coordinates and checked it. The line was just in the inlet. I also tried to rotate the last profile file and the coordinates were the same as the first one. So the coordinate transformation worked...

sbaffini · August 17, 2017, 12:33

If everything is correct, the nearest neighbor interpolation should simply work

August 7, 2017, 10:20	Profile Boundary Condition	#1
sufjanst Member Join Date: Mar 2016 Posts: 73 Rep Power: 10	Hello, I try to simulate a combustion problem. I already simulated a domain with 4 burners. Now I'm trying to simulate a domain with 8 burners which are the same as in the first domain but on other positions. I created a Hexa-Mesh to have a complete symmetric domain. I wrote a profile for one burner and tried to use it as a boco for each of the 8 burners. This didn't work. Is that even possible due to the fact that the mesh positions are different at the 8 burners? Is there a solution for this problem? Here are the positions for the 8 burners. https://picload.org/view/rpppapgw/2.png.html Sufjanst

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
My radial inflow turbine	Abo Anas	CFX	27	May 11, 2018 01:44
Out File does not show Imbalance in %	Mmaragann	CFX	5	January 20, 2017 10:20
Error - Solar absorber - Solar Thermal Radiation	MichaelK	CFX	12	September 1, 2016 05:15
Velocity profile boundary condition	Tuca	FLOW-3D	1	April 23, 2013 12:02
velocity profile inlet boundary condition question	Lcw	FLUENT	3	August 3, 2012 05:53

August 7, 2017, 10:28		#2
sbaffini Senior Member Paolo Lampitella Join Date: Mar 2009 Location: Italy Posts: 2,151 Blog Entries: 29 Rep Power: 39	If by profile you mean a profile file for the inlet, than I'm more than 99% sure that it is at least one of your problems. Those files are meant to be used as interpolation tables, which depend on the coordinates of course. Your best option would be to manually rotate your original domain and produce a new file for the new position, which will superpose with the rest of the final domain, then use both profile files. Otherwise, if you are into it, you can write a small script (say, Octave/MATLAB) that will read the file, rotate the coordinates, and write it back. I don't know if recent Fluent editions have a feature for directly doing this on the files... I guess not.

August 14, 2017, 07:10		#3
sufjanst Member Join Date: Mar 2016 Posts: 73 Rep Power: 10	thank you for your reply. I wrote a Matlab script, which rotated my coordinates and profiles. I used them as inlets and they work! I now have a problem that a few cells have very high velocities. All of the inlets have a velocity magnitude of about 20 m/s. Except for 10 cells which have a velocity magnitude of 1044514 m/s. I don't know why. I tried every interpolation method...

August 15, 2017, 16:13		#4
sbaffini Senior Member Paolo Lampitella Join Date: Mar 2009 Location: Italy Posts: 2,151 Blog Entries: 29 Rep Power: 39	How did you check that your script to rotate the coordinates was correct? Looks like it might have a singularity for some coordinates. Can you post it here?

August 16, 2017, 16:59		#5
sufjanst Member Join Date: Mar 2016 Posts: 73 Rep Power: 10	The coordinates are correct. I created a line with two coordinates and checked it. The line was just in the inlet. I also tried to rotate the last profile file and the coordinates were the same as the first one. So the coordinate transformation worked...

August 17, 2017, 12:33		#6
sbaffini Senior Member Paolo Lampitella Join Date: Mar 2009 Location: Italy Posts: 2,151 Blog Entries: 29 Rep Power: 39	If everything is correct, the nearest neighbor interpolation should simply work