Fluids not passing through gaps.


#1

Hi all, I am working on a scene that needs mist to come up through the gaps in the geometry in the floor. The problem I am having, is that the fluid wont pass through these gaps, but instead it is just colliding with the base of the geometry and is only passing through the big gap in the middle.

Here is an example of what i mean.

http://www.youtube.com/watch?v=Gm0TXHA33MU&feature=youtu.be

Any help on this matter would be much appreciated.

Cheers, Kai.


#2

Either your model is not correctly cut, or your fluid res is too low res.


#3

mmm well try with more resolution


#4

You can set shadedDisplay to collision to see the voxels that are marked in collision( the mesh is voxelized and the collision is with these voxels, not the exact mesh). Also note that the collisions are a bit more accurate if you use forward advection.


#5

but is it then with forward advection and a more accurate collision detection even slower ?

ohh and mr duncan - i tried to send you a pm but your inbox is fulll. :wink:


#6

Forward advection is a bit slower, but by how much depends on your simulation. It might not make much difference in the speed. It is more leak proof then the default backward propagation, which can allow fluid to cross boundaries if the velocity is high enough. Also it does not zero density in occupied voxels, and thus also better preserves total density.

The resolution for collision is still voxel based, so the resolution is no different, however it does handle moving collision boundaries better.


#7

yeah i know. so to have exact collision we have use very high resolution for now.

could it be speed up and make it more detailed when we convert the collison objects
to nCloth and use a bit force field generation with a small distance and not the voxelbased collision - so we are independent from the voxel resolution ?


#8

In some cases you might get smoother results by using speed fluid emission (density emission off) from a surface with the replace method and inherit set to 1.0, however it is not a true collision, just a zeroing of velocity as one gets close the the surface. One can use the maxDistance on the fluid emitter as well as the fluid dropoff to create a smooth transition. It is a blurrier boundary then the collisions provide.


#9

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.