[sharonyue]

Originally Posted by Ya_Squall2010

Please see attached image. There is indeed a gap in the patches but only visible after zoomed in for many many many times. The gap in the results, however, is much much more bigger than the physical one. Also, from the other picture where the view angle is from below the free surface, we can easily see a thin layer of water is "leaking" in between the AMI gap!

Attachment 21570
Attachment 21571

Hi,
I just run my case,and I found the key of the gap.See this two pictures.its the same result.but by different launched method paraview.(Sorry for my english)
The first one's result is opened by file-open-xxx.foam(or launched by command "paraview").The second one is opening xxx.OpenFOAM(or launched by command "paraFoam").Although theres a small gap in second one.I think it doesnot matter.
The third picture is only showing AMI patch.I think its fine.



See the gap in this two picture.I think its just a visual problem which would not affact the result.

[Ya_Squall2010]

Can you spot "leaking" by checking from below?

[sharonyue]

Originally Posted by Ya_Squall2010

Can you spot "leaking" by checking from below?

There was no leaking in my case,at least in clip ot contour.
How do you know there is a "leaking"??

[Ya_Squall2010]

as can be seen from one of my previous posted images, where the iso surface of water (vof=0.5) is shown, the water flows in between the gap as the arrows pointed out. I am not sure whether this is due to paraview interpolation. I also found for AMI+vof cases, the point interpolated vof, unlike the original cell centred vof, has higher value than 1 in paraview. Could you please check yours as well?

[sharonyue]

Originally Posted by Ya_Squall2010

as can be seen from one of my previous posted images, where the iso surface of water (vof=0.5) is shown, the water flows in between the gap as the arrows pointed out. I am not sure whether this is due to paraview interpolation. I also found for AMI+vof cases, the point interpolated vof, unlike the original cell centred vof, has higher value than 1 in paraview. Could you please check yours as well?

Yep,in my case its larger than 1.and its in the interface of AMI.but after use cell center it returns back to 1.I dont know why its this.Do you know why?I look forward to you reply~Thanks.


Oh, I check my case no matter how I use filter I cant see leaking.

[jrrygg]

Hi there,

I think I have a similar problem as you Ya_Squall2010, please see the attached picture. Only my isovolume is shown (0.5<alpha<1.01), and you can se that the water jet crossing the AMI-patches has a gap where the interface is, and also a slight leak (upper part of the picture). The leak starts around the jet, and seems to follow the outside of the AMI-patch as it rotates. I am not sure if the "leak" is trapped between the AMI-patches or stays in the outside/right domain along the AMI-patch. To me the latter seems to be the case.

Also the alpha-values fluctuate between a (very small) negative value and a value larger than 1, while it should ideally be bounded between 0 and 1. I am not sure if this can be avoided, but I don't think it is really a problem.

From the log:

AMI: Creating addressing and weights between 13636 source faces and 13636 target faces
AMI: Patch source weights min/max/average = 0.9998931035, 1.000192161, 1.000016132
AMI: Patch target weights min/max/average = 0.9998635359, 1.000282629, 1.00001155
Execution time for mesh.update() = 0.13 s
MULES: Solving for alpha1
Phase-1 volume fraction = 0.008494118373  Min(alpha1) = -5.55245485e-17  Max(alpha1) = 1.000022429

If I understand it correctly the AMI: Patch source weights values mean that the AMI-patches are not 100 % aligned, i.e. not perfectly circular and equal (but very close to it).

It think especially the "leak" behavior is very strange, do you have any idea about the reason of this? Also, it seems that the gap in the isovolume is larger than the actual gap between the AMI-patches, however this might just be visual.

Regards,

Jone

[sharonyue]

Hi guys,

I tried another case in which the geometry size is bigger than my last one. So there are more fluctuated isosurface. In this case I think its good.because the liquid passing through AMI is connected even the liquid is flying...But I cannot spot leaking.
and I dont know if the leaking thing is normal.

[Ya_Squall2010]

you need to check the leaking from below.

[Ya_Squall2010]

Hi, jrrygg,

The vof range in your log file seems fine to me. But what is the range when you do post-processing in paraview?

Yes, I can see fluid got trapped between the AMI patches in your image as well. I have no idea why this is happening, and what is the effect to the overall accuracy. Let's hope this is just due to the postprocessing interpolation as non-conformal AMI patches are two seperated entities to paraview anyway, and the later doesn't know how to produce a smoothed solution across them.

Best luck.

[jrrygg]

Originally Posted by Ya_Squall2010

The vof range in your log file seems fine to me. But what is the range when you do post-processing in paraview?

Thank you for your reply. The range is the same in Paraview, meaning that I have to specify an isovolume between 0.5 (or the desired minimum) to 1.0001 or something similar to get an appropriate view of the water. If I use 1 as a maximum instead large parts of the water jet (mainly around the inlet) will disappear.

Originally Posted by Ya_Squall2010

Yes, I can see fluid got trapped between the AMI patches in your image as well. I have no idea why this is happening, and what is the effect to the overall accuracy. Let's hope this is just due to the postprocessing interpolation as non-conformal AMI patches are two seperated entities to paraview anyway, and the later doesn't know how to produce a smoothed solution across them.

Is there any way we can test this effect without using paraview? I will of course validate my results, but I don't know a good way to isolate this specific problem.