Ryan Goldade

I would recommend enabling “Apply Moving Collisions”, use a “Min Air Region Voxels” of something like “5”. Without moving collisions, your bubbles are pulled along with the cup so holes open up. The Min Air Region allows small bubbles to collapse, so if bubbles do form along the cup as it's moving they can hopefully collapse back down. Finally, adding more substeps could help (although I still got decent results without changing the substeps).

I took a look at the boat. It seems like the resolution of the flip simulation is not high enough to capture the boat. One thing to keep in mind is that the size of a voxel in the flip simulation is (particle separation) * (grid scale). If the canoe is not thicker than one or two voxels, it won't be accounted in the solver. A good way to check is to enable “Collision” in the Guides->Visualization tab in the FLIP Object node. It will show a purple mesh of the collision volume.

I had to decrease the particle separation to .025 before I could even see the canoe show up as a collision. I imagine you will need to make it much smaller for the sides of the canoe to show up. This will definitely take longer to simulate so it might be preferable to reduce the size of the tank that you're using.

Are you using the same rigid body option for the boat as the torus? Also, how thick is your boat? If the shell of the boat is too thin, then the flip solver might not realise the boat as a collision (depending on the particle separation of course).