CFX rotating Domain

cfxbeginner · June 16, 2017, 06:49

I have a computer problem so I will do it this afternoon.
I have read a Training manual and find that if we create a rotating domain the CFX-Solver computes the appropriate Coriolis and centrifugal momentum terms.
Could it be my problem?
Here is the text

Training_manual.PNG

ghorrocks · June 16, 2017, 07:01

No, it should not be.

Go back to the definition of a frame of reference is. It is simply something to measure distances and velocities relative to. While the velocity is different values in different frames of reference, the velocity in the absolute frame of reference is unchanged. To change the velocity you need something physical - a wall, pressure gradient or something like that.

So the test you have done is a good check that simply adding a rotating frame of reference does not change the flow in the absolute frame. The fact that it appears that it is changed means something is wrong.

cfxbeginner · June 19, 2017, 07:05

I simplified a bit my problem and stop the inlet velocity. The velocity near the interface seems not to be null and the velocity in the rotating domain is smaller but the fluid seems to rotate.
Are These perturbations due to GGI parameters or my mesh? What can I do to improve it?

velocity_pb-sscylinder-ssinlet_2.jpg

velocity_pb-sscylinder-ssinlet_3.PNG

urosgrivc · June 19, 2017, 07:45

Do you think that this is the cause?
Velocity here has a magnitude something to e10-4

Results vhich are problematic described in previous page are in orders of few 100% wrong ,that is wierd i think that this is negligable compared to those figures.

cfxbeginner · June 19, 2017, 10:24

I don't see any other explanations. Is it possible that the fluid velocity increase these little errors? I really don't understand what is my mistake ...
The solution with an hexa-mesh seems to be worst... I'm now trying to improve my mesh.

LorenzoSaintDubois · June 21, 2017, 11:00

Hello,

did you take a look at the imbalance values? The CCL that was posted did not include them under the convergence criteria... It can be that the residual convergence is tight but the imbalances are completely off, so the physical solution doesn't make sense.

Good luck,

LSD

June 21, 2017, 11:00	Imbalances	#26
LorenzoSaintDubois New Member Join Date: Mar 2015 Posts: 5 Rep Power: 11	Hello, did you take a look at the imbalance values? The CCL that was posted did not include them under the convergence criteria... It can be that the residual convergence is tight but the imbalances are completely off, so the physical solution doesn't make sense. Good luck, LSD

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Rotating or stationary domain in pump/turbine cavity	Christian Høy	CFX	3	March 17, 2017 06:03
convergenceof natural convection prob. in cfx	cpkewat	CFX	15	January 31, 2014 06:29
injection problem	Mark New	FLUENT	0	August 4, 2013 01:30
Different conductivities for a SINGLE domain for CHT problem in CFX	Mehul	CFX	3	August 3, 2012 09:01
question about governing equation in CFX using rotating/non rotating reference frame	rystokes	CFX	0	January 12, 2010 06:14

June 16, 2017, 06:49		#21
cfxbeginner New Member Jules Join Date: Jun 2017 Posts: 12 Rep Power: 8	I have a computer problem so I will do it this afternoon. I have read a Training manual and find that if we create a rotating domain the CFX-Solver computes the appropriate Coriolis and centrifugal momentum terms. Could it be my problem? Here is the text Training_manual.PNG

June 16, 2017, 07:01		#22
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	No, it should not be. Go back to the definition of a frame of reference is. It is simply something to measure distances and velocities relative to. While the velocity is different values in different frames of reference, the velocity in the absolute frame of reference is unchanged. To change the velocity you need something physical - a wall, pressure gradient or something like that. So the test you have done is a good check that simply adding a rotating frame of reference does not change the flow in the absolute frame. The fact that it appears that it is changed means something is wrong.

June 19, 2017, 07:05		#23
cfxbeginner New Member Jules Join Date: Jun 2017 Posts: 12 Rep Power: 8	I simplified a bit my problem and stop the inlet velocity. The velocity near the interface seems not to be null and the velocity in the rotating domain is smaller but the fluid seems to rotate. Are These perturbations due to GGI parameters or my mesh? What can I do to improve it? velocity_pb-sscylinder-ssinlet_2.jpg velocity_pb-sscylinder-ssinlet_3.PNG

June 19, 2017, 07:45		#24
urosgrivc Senior Member urosgrivc Join Date: Dec 2015 Location: Slovenija Posts: 365 Rep Power: 11	Do you think that this is the cause? Velocity here has a magnitude something to e10-4 Results vhich are problematic described in previous page are in orders of few 100% wrong ,that is wierd i think that this is negligable compared to those figures.

June 19, 2017, 10:24		#25
cfxbeginner New Member Jules Join Date: Jun 2017 Posts: 12 Rep Power: 8	I don't see any other explanations. Is it possible that the fluid velocity increase these little errors? I really don't understand what is my mistake ... The solution with an hexa-mesh seems to be worst... I'm now trying to improve my mesh.