Fluid ignores animation of static object [SOLVED]

[TheoK]

Hello,

 

This is my first ever post here. Usually I try to fix things by myself since it helps me learn, but I've been having this problem for almost 2 weeks now and it's driving me insane.

 

I have modelled a storage shelf and placed buckets on it. The shelf gets hit by a ball and all the buckets fall on the ground. Some of the buckets have paint in them which is supposed to stay inside until the lids open.

I have simulated the buckets falling and I have cached the simulation.

I'm currently testing the fluid sim with only one bucket and the problem I'm having is that the fluid collides with the bucket just fine when it gets created, but when the animation of the bucket starts, the fluid just stays in its place as if

the bucket never moved. Once the bucket falls out of place, the fluid also just falls down completely ignoring the bucket's movement.

 

I have checked my collision geometries and they all look fine and follow the animation. I am using volume collisions, I have tried filling their interiors, extruding the collision geometry by a lot, increased the substeps, resizing the fluid geo and every time I get the same result.

So I don't think it's a problem with the collision geometry itself, but more with how the fluid reacts to the animation of the static object. It almost looks like as if it's colliding with a bucket with no animation, which suddenly disappears.

 

I am pretty sure the solution is something simple that I have failed to notice but I can't figure out what it is for the life of me.

 

Any help would be greatly appreciated.

 

Thank you.

Edited March 22, 2021 by TheoK
solved

[TheoK]

Anyone had this problem before or has the slightest idea as to why it's happening?

[vicvvsh]

1 hour ago, TheoK said:

Anyone had this problem before or has the slightest idea as to why it's happening?

Hi,

you can explore this hip. Increase buckets' sides depth, use 'Collision Object' shelf tool for them to create collision shapes, increase substeps on flip solver.

 

flip.hipnc

[TheoK]

Thank you for the reply @vicvvsh

I'll check it now.

[TheoK]

Hi.

I checked that file and explored it as much as I could. I even integrated it on mine and vice versa and still couldn't make it work.

I also tried your suggestions. I have already tried increasing the buckets' sides depth and increasing the substeps on the flip solver and I used the static object shelf tool but still the same result.

 

I'm not 100% sure what the problem is but I noticed that when I visualize the collision guide on the flip object, the geometry is not the same as the static object.

I have imported an exaggerated extruded geometry as a static object, just to show you what I mean.

 

In the first image, you can see the static object in grey and its volume collision geo in blue.

In the second image, I have visualised the collision guide on the flip object and apparently it's different and looks like the bucket without any extrusion.

 

I tried adding a new flip object and it's still the same. So it seems like the flip object is not using the static object as a collision but I don't understand why it's using this object or how to fix it.

[TheoK]

Ok, so that was happening because I had Invert Sign checked on the static object.

So that's not an issue anymore, but the main problem still exists.

[TheoK]

OK, so I figured it out finally.

Thanks to your file @vicvvsh, I realised that in the flipsolver, I had left the Velocity Scale of the collisions to 0.

That's why the fluid was interacting with the collision geo but wasn't following it when it moved. Which makes a lot of sense and I thought I had already tested that, but apparently I hadn't.

I put it to 1 and all is working as it should now.

Thank you again for your help. Hopefully that will help someone else too.

Edited March 14, 2021 by TheoK
Added image