[k323r]

Dear FOAMers,

I've been working on dynamic meshes for quite a while now, trying to implement insect wing movement. Sucess is rather limited so far. First i tried for quite a while by using dynamicTopoFvMesh; while the library is a neat little tool, mesh quality is rather a bummer. So now i'm trying to implement the movement by embedding three AMI cylinders into each other (one for each rotational axis), though that's not really working either.

So has anyone in the community ever tried such a (probably stupid) thing? The problem is that, while all the normal patches get dragged along, the slave-patches (containing the acutal wing) stay put behind (see screenshot in the appendix, the third cylinder, rotating around the z axis is not shown).

I'd be super grateful for any input or ideas on the matter. Basically what i'm trying to accomplish is the rotation of a patch around all three axis with quite large amplitudes.

Thanks in advance,

Josch

[micpage18]

Hi Josch,

What you need is to have access to immersed boundary condition. Official OpenFoam doesn't support this feature actually. However, a toolkit can be used with the Extended version.

https://openfoamwiki.net/index.php/E...mersedBoundary

Keep us informed if you try it.

Good luck

[k323r]

Hi Michael,

thanks for your reply! i'll try it right away - at the moment i'm running into a few compilation errors, but i'll keep you posted. As to the wing motion problem: i realised, that the simplest way to model the movement is by embedding the wing model in a sphere and describe the rotation by the wings surface normal vector - so at the moment that seems to be another approach to the problem.

again, thank you for your help.

cheers,

josch

[louisgag]

Hi Josch,

If you're looking for an implementation for an embed AMI interface you could start from my code: http://www.cfd-online.com/Forums/ope...tml#post476869

You would have to adapt it to include a third level AMI and modify the normal vectors of the outer rotations because my rotations were all in the same plane.

-Louis

[k323r]

Hi Louis,

neat code! just briefly went through it, looks good thanks for the hint! i'll test it, compilation worked like a charme! I assumed by 'in the same place' your rotations all had the same centre of origin?

[louisgag]

Hi Josch,
Glad it can be useful.
I wrote in the same plane because all my rotating zones have their rotation axis normal to the x-y plane)
I think the modifications required should not be that complicated.
Let me know about it!
-Louis