Moving coordinate frame

Roland R · November 23, 2011, 12:51

Hello,

I would like to rotate my subdomain with deforming mesh. This is OK. But this domain moves in a straight line too. Rotation of the body is based on the original X-Y-Z coordinate frame but of course it should move with the body together.
How can I define moving coord frame?

Regards
Roland

singer1812 · November 23, 2011, 15:16

This is very easy to do, particularly if you are deforming the whole subdomain (then remeshing isn't necessary). If you use rotation and translation matricies to determine your movement you can combine the translation and rotation into one equation, and it can be as complicated as you wish it to be. If you remesh, you will need to add in a de-rotation (of the rotation at the remesh time) and a de-translation (at the time of the remesh time) matricies prior to the actual rotation and translation.

If you give more details on how you are trying to move your mesh, perhaps I can be more specific in an example.

Roland R · November 28, 2011, 05:42

Hello Singer,

thanks for your answer, but could you give me more detailed explanation? To tell the truth, I don't understand quite...

My problem:
I would like to simulate a simple 3D aircraft during a genaral figure looping. In addition strong wind attacks the aircraft body therefore the original circle figure will deform. I would like to analyse this deformed looping with combination of translation and rotation.

Regards;
Roland

Quote:

Originally Posted by singer1812

This is very easy to do, particularly if you are deforming the whole subdomain (then remeshing isn't necessary). If you use rotation and translation matricies to determine your movement you can combine the translation and rotation into one equation, and it can be as complicated as you wish it to be. If you remesh, you will need to add in a de-rotation (of the rotation at the remesh time) and a de-translation (at the time of the remesh time) matricies prior to the actual rotation and translation.

If you give more details on how you are trying to move your mesh, perhaps I can be more specific in an example.

singer1812 · November 28, 2011, 10:59

By figure looping do you mean something like this?

And if so, couldnt the translation be taken up via the inlet velocity so that you would only be dealing with pure rotation of the body?

November 23, 2011, 12:51	Moving coordinate frame	#1
Roland R Senior Member Roland Rakos Join Date: Mar 2009 Posts: 131 Rep Power: 17	Hello, I would like to rotate my subdomain with deforming mesh. This is OK. But this domain moves in a straight line too. Rotation of the body is based on the original X-Y-Z coordinate frame but of course it should move with the body together. How can I define moving coord frame? Regards Roland

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CFD Post- coordinate frame issue	GSbert	ANSYS	1	June 15, 2012 02:26
Moving refence frame	yagizb	FLUENT	0	March 19, 2011 04:14
does sb can tell me what is the differences between moving mesh and moving frame?	caohan	FLUENT	0	June 2, 2010 06:15
Help for Moving Reference Frame.	aiqch	Main CFD Forum	0	April 3, 2009 01:23
Moving reference frame	CFD_boy	FLUENT	6	April 20, 2006 12:39

November 23, 2011, 15:16		#2
singer1812 Senior Member Edmund Singer P.E. Join Date: Aug 2010 Location: Minneapolis, MN Posts: 511 Rep Power: 21	This is very easy to do, particularly if you are deforming the whole subdomain (then remeshing isn't necessary). If you use rotation and translation matricies to determine your movement you can combine the translation and rotation into one equation, and it can be as complicated as you wish it to be. If you remesh, you will need to add in a de-rotation (of the rotation at the remesh time) and a de-translation (at the time of the remesh time) matricies prior to the actual rotation and translation. If you give more details on how you are trying to move your mesh, perhaps I can be more specific in an example.

November 28, 2011, 10:59		#4
singer1812 Senior Member Edmund Singer P.E. Join Date: Aug 2010 Location: Minneapolis, MN Posts: 511 Rep Power: 21	By figure looping do you mean something like this? And if so, couldnt the translation be taken up via the inlet velocity so that you would only be dealing with pure rotation of the body?