CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Pressure fluctuations along transient rotor-stator interface (http://www.cfd-online.com/Forums/cfx/74311-pressure-fluctuations-along-transient-rotor-stator-interface.html)

Norbert March 30, 2010 02:53

Pressure fluctuations along transient rotor-stator interface
 
Hi there,

I have a test case consisting of two concentric cylinders, which are connected to each other with transient rotor-stator interface on a cylindrical face.
I use this interface because later on I would like to run sliding mesh in an axial fan simulation.
The inner and outer surface mesh size on the interface is appr. 1:2 with a resolution of appr. 0.6 deg/cell in the circumferential direction and appr. 1.3 deg/cell for the outer surface.
The resulted pressure field shows stripes parallel to the rotation axis along the interface in a magnitude that cannot be neglected compared to the average static pressure.
I think this is related to the mesh density on the surface as with a higher density the stripes were not present.
I saw similar fluctuations in fluent earlier when the interfaces were not close enough or overlapped.
I saw that there are several GGI options to modify the interface behavior and already tried to increase the bitmap resolution from 100 to 200 without any luck.
Have you experienced similar behavior before? Is it possible to fix this without increasing the mesh density?

Norbert

ghorrocks March 30, 2010 18:23

Can you post a drawing and describe what the flow is doing?

Norbert March 31, 2010 02:46

1 Attachment(s)
I attach a sketch containing the domain and a pressure plot at the interface.
I hope this explains more clearly.

Attachment 2752

ghorrocks March 31, 2010 07:12

Sounds unusual. I guess the GGI is having troubles connecting but you already guessed that. I guess your options are, as you say, either a finer mesh or to fiddle with the GGI parameters. But you already knew that.

You may also get better behaviour if you split the exit GGI into a two GGIs, one radial and one longitudinal. Just a guess but it may behave better like that.

Norbert March 31, 2010 07:54

I have already tried to separate the exit GGI in the way you mentioned.
Actually it seems that faces that are perpendicular and parallel to the rotation axis are not allowed at the same transient rotor-stator interfaces. So I had to separate it into two regions anyway. I will play around with the other GGI parameters.


All times are GMT -4. The time now is 15:25.