Export from post gives different result between v11 and v12

bennn · February 9, 2011, 05:29

hey everyone

We just installed v12, and we have something pretty big.

Our problem is a darrieus rotor, basically a 2D domain with a blade rotating around a center of rotation away from it. http://commons.wikimedia.org/wiki/File

arrieus.jpg

The output we're looking at is the aerodynamic forces on each blade during the rotation, given by force_x(or y)@FOIL.

We had a .res simulated on v11. The thing is when exporting the force data from v11 post and then from v12 post, the results are different... The projection itself (force_x and force_y) are respectively on rotating ref frame and fixed ref frame (which can be understood from the help files), but the values themselves after projecting adequately are different from one version to the other. it's a post issue only. there's only one .res file there so it's just how the post interprets the results.

The help files on what force_ calculates are strictly the same. I suspect a force component being added in one case, but I can't find a way to calculate each component separately (unlike fluent).

That's really surprising from such a big software... and pretty scary actually for someone right in the middle of a phD.

let me know what your thoughts are.

Thanks !!

ghorrocks · February 9, 2011, 06:42

Could be a bug.

I would recommend doing a flow in a simple flat plate. That way you can work out the pressure and viscous forces pretty easily and compare it to force_x/y/z. Might as well have a look at torque functions while you are at it.

Also compare the force_x/y/z from CFD-Post to that given by a monitor point in the solver. Is it the same?

If you can show the current CFX version is wrong then you have a bug so report it and get it fixed. If you can show the older version of CFX is wrong.... well, time to upgrade.

bennn · February 10, 2011, 06:38

ok so the forces from v12 solver's monitor are the same, except they are in the rotating frame. but when projected it's the same as v12 post.

we're tlaking with ansys right now to get to know if it'as a bug or not.

thanks for your help

February 9, 2011, 05:29	Export from post gives different result between v11 and v12	#1
bennn Member benoit paillard Join Date: Mar 2010 Posts: 96 Rep Power: 16	hey everyone We just installed v12, and we have something pretty big. Our problem is a darrieus rotor, basically a 2D domain with a blade rotating around a center of rotation away from it. http://commons.wikimedia.org/wiki/Filearrieus.jpg The output we're looking at is the aerodynamic forces on each blade during the rotation, given by force_x(or y)@FOIL. We had a .res simulated on v11. The thing is when exporting the force data from v11 post and then from v12 post, the results are different... The projection itself (force_x and force_y) are respectively on rotating ref frame and fixed ref frame (which can be understood from the help files), but the values themselves after projecting adequately are different from one version to the other. it's a post issue only. there's only one .res file there so it's just how the post interprets the results. The help files on what force_ calculates are strictly the same. I suspect a force component being added in one case, but I can't find a way to calculate each component separately (unlike fluent). That's really surprising from such a big software... and pretty scary actually for someone right in the middle of a phD. let me know what your thoughts are. Thanks !!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Problem Interface Solid Fluid with wall velocity Solver v12	hills1	CFX	2	October 12, 2009 06:36
can you import the cmdb (workbench v11) file into CFX v12?	MHZ	CFX	0	April 15, 2009 16:03

February 9, 2011, 06:42		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	Could be a bug. I would recommend doing a flow in a simple flat plate. That way you can work out the pressure and viscous forces pretty easily and compare it to force_x/y/z. Might as well have a look at torque functions while you are at it. Also compare the force_x/y/z from CFD-Post to that given by a monitor point in the solver. Is it the same? If you can show the current CFX version is wrong then you have a bug so report it and get it fixed. If you can show the older version of CFX is wrong.... well, time to upgrade.

February 10, 2011, 06:38		#3
bennn Member benoit paillard Join Date: Mar 2010 Posts: 96 Rep Power: 16	ok so the forces from v12 solver's monitor are the same, except they are in the rotating frame. but when projected it's the same as v12 post. we're tlaking with ansys right now to get to know if it'as a bug or not. thanks for your help