CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   STAR-CCM+ (https://www.cfd-online.com/Forums/star-ccm/)
-   -   DFBI - Whole Domain Falling (https://www.cfd-online.com/Forums/star-ccm/241114-dfbi-whole-domain-falling.html)

DRiver February 9, 2022 07:35

DFBI - Whole Domain Falling
 
1 Attachment(s)
Hello, I am running a simulation about ship resistance including the DFBI model for rotation and translation. I have activated the Z translation (Sinkage) and Y rotation (Trim) in the free motion model. When running, the body release ramp starts at 1.0 s and it ends at 5.0 s. My simulations start correctly but after the time 2.0 s, the whole domain starts to fall. I attach the GIF of the Isometric scalar scene to show what I mean. Based on the Position Z [m] color scale, you can see how absurd values this variable reaches. You can also see how the outline of the domain disappears, therefore, the domain is also falling.

I make clear that the body in the DFBI tree is only composed of the Deck and Hull surfaces.

Has any of you experienced this situation? Have you found the source of the error or any hint to solve this issue? :confused:

Thank you.

https://media.giphy.com/media/mQ9pUH...iFnO/giphy.gif

xkevbear June 6, 2022 09:55

I also need help with this. However, my domain stayed but my ship sanked.

koful June 10, 2022 10:53

Hello,

DFBI is a region based model (edit: I am not very sure of this sentence). If you have only one region, the motion will effect the regions boundary. So, the falling of the entire region is funny but correct. Maybe you over-weighted the DFBI object? High denisty can cause sinkage.

Aeronautics El. K. June 25, 2022 04:13

Quote:

Originally Posted by koful (Post 829626)
Hello,

DFBI is a region based model (edit: I am not very sure of this sentence). If you have only one region, the motion will effect the regions boundary. So, the falling of the entire region is funny but correct. Maybe you over-weighted the DFBI object? High denisty can cause sinkage.

That's not quite right. I use the DFBI with 2DoF routinely with 1 region and both the region and the ship stay were they should.

I agree about checking the weight of the vessel but there seems to be more to it than that in DRiver's setup.

koful June 27, 2022 02:55

Quote:

Originally Posted by Aeronautics El. K. (Post 830445)
That's not quite right. I use the DFBI with 2DoF routinely with 1 region and both the region and the ship stay were they should.

I agree about checking the weight of the vessel but there seems to be more to it than that in DRiver's setup.

Hi,

Thank you for addition.

If there is an overset region it travels inside the background mesh. If there is only one region the region still travels. If you use only 2dof, of course the ship will not move towards the x direction.

You can check this video and you can see that whole region is rotationg and translating.
https://www.youtube.com/watch?v=jFtZKNMVrc0

I may wrong, I will check the documentation.

Aeronautics El. K. June 27, 2022 03:54

Quote:

Originally Posted by koful (Post 830526)
Hi,

Thank you for addition.

If there is an overset region it travels inside the background mesh. If there is only one region the region still travels. If you use only 2dof, of course the ship will not move towards the x direction.

You can check this video and you can see that whole region is rotationg and translating.
https://www.youtube.com/watch?v=jFtZKNMVrc0

I may wrong, I will check the documentation.

True, the overset will move in the background domain depending on how many (and which) degrees of freedom are allowed.

In my setup I don't use overset, it's not needed. I use morphing mesh. And I don't quite need the movement in the x direction, I only allow heave and trim.
That's perfectly doable with 2 dof, 1 region and morphing mesh.

I'm still curious why the domain moves in the original post. It would be good if we had more details about the setup.


All times are GMT -4. The time now is 20:50.