[SSSERB13]

Hello mates,

I have a question about the results that are appearing after the solution of the rotating movement. I am running a VAWT using pimpleFoam and AMI to count the transient effects. We can see clearly the velocity gradient on the AMI interfaces, however, the case is running well. Is there any way to build a more smooth gradient between the rotating and stationary domain?

Please find attached the image that illustrates the velocity magnitude results.

Really appreciate your help.

Regards;

[simrego]

Hi!


I think your a huge jump in cell size on the two sides of the AMI, try to reduce it.
Also check your AMI BCs because I think there are some problems... You should see nearly nothing between the stationary and moving mesh.

[SSSERB13]

Originally Posted by simrego

Hi!


I think your a huge jump in cell size on the two sides of the AMI, try to reduce it.
Also check your AMI BCs because I think there are some problems... You should see nearly nothing between the stationary and moving mesh.

Hi @simrego,
The size transitioning of my mesh is not bad as you can see in the image attached.
I checked the AMIS BCs and looks ok too. There are the dynamicMeshDict, U, p and turbulence that were reviewed too. Do you have a specific tip about the files and parameters?
When dealing with moving frames, there are some options of what type of velocity you want to see, if it is in a rotating or stationary frame. Do you know how we can access these two kinds of fields using OF? Maybe could be due to this.

Thank you for your attention. Very appreciated.

[simrego]

Yes. From that first plot it seemed to be much much worse...



If you are simply using a rotation in the dynamicMeshDict, you will get the results in the stationary frame for both meshes.


Without any specific options it'll be mostly a guessing game from everyone. Try to share some files because from these pictures it can be nearly anything.
But I still think that this AMI BC is wrong somewhere. It seems like your inner and outer meshes are not coupled correctly. From this figure, it seems like nothing happens inside the inner mesh like if it is a closed domain...
You can just try to turn off the rotation, and simply add some inlet velocity on the side and check the flow field. You will probably see the problem.

[SSSERB13]

Originally Posted by simrego

Yes. From that first plot it seemed to be much much worse...



If you are simply using a rotation in the dynamicMeshDict, you will get the results in the stationary frame for both meshes.


Without any specific options it'll be mostly a guessing game from everyone. Try to share some files because from these pictures it can be nearly anything.
But I still think that this AMI BC is wrong somewhere. It seems like your inner and outer meshes are not coupled correctly. From this figure, it seems like nothing happens inside the inner mesh like if it is a closed domain...
You can just try to turn off the rotation, and simply add some inlet velocity on the side and check the flow field. You will probably see the problem.

Yes, should be something wrong but I am struggling to find out. In the new image attached, it's possible to see the plot over the line and the velocity descontinuity. When putting off the rotation the flow develops normally.
I also shared the BCs files, if you could check I would appreciate it.
Thank you for your help!