Hello all,
first let me tell you how happy I am with Houdini. I wish I had discovered it
earlier ops: Great Job SideFX.
So I´m using Houdini now for roughly a week and went through several
Tutorials. So I´m pretty much at the very beginning.
I ´m trying to make a very basic simulation. A Cup that ´s being filled
with liquid by a Particle Emitter. I build a cup from a cylinder and made
a RBD from it. I place a sphere above as emitter for my FLIP fluids.
The particles pour in but somehow the Cup doesn´t fill. The particles
disappear. I allready tried the following without success.
Increasing the “Uniform Divisons” from the Cup Object.
Using a static solver instead of a rigid body solver (as seen in the screenshot).
As you can see in the attached image, it´s not filling. It looks pretty much
the same after 200 frames.
I also have the problem that some particles just pass through the Cup.
It gets better when I increase the uniform subdivisions. But maybe
there are more usefull parameter to suspress this behaviour ?
Any help or pointer to a tutorial is much appreciated.
Cup not filling with FLIP Fluids
13947 5 1- netfrag-sam
- Member
- 12 posts
- Joined: 6月 2012
- Offline
- br1
- Member
- 41 posts
- Joined: 3月 2008
- Offline
- netfrag-sam
- Member
- 12 posts
- Joined: 6月 2012
- Offline
Yes it seems it ´s a problem with this particular geometry.
Actually the geometry looks correct. The Cup has a thickness
and I can see no intersections. Also the collision geometry looks correct
when I turn on “show collision guide geometry”.
I started a new scene used a box instead of a cylinder and it works
now as expected. So for the moment I´m ok with that as I´m just testing.
Thank You !
Actually the geometry looks correct. The Cup has a thickness
and I can see no intersections. Also the collision geometry looks correct
when I turn on “show collision guide geometry”.
I started a new scene used a box instead of a cylinder and it works
now as expected. So for the moment I´m ok with that as I´m just testing.
Thank You !
- eitht
- Member
- 257 posts
- Joined: 11月 2010
- Offline
Hello, i'm fairly new to FLIP too.
To fill up the cup fast, first thing that comes to my mind is to increase the number of points on the source emitter.
For particles penetrating the cup, I would suggest using the cup itself as a sink to kill the naughty particles. Append Peak SOP to your cup geometry and scale it up and use that as a sink.
Think you can use Delete SOP too but it doesn't take in a bounding object like Group SOP so may not be a good idea.
eitht.
To fill up the cup fast, first thing that comes to my mind is to increase the number of points on the source emitter.
For particles penetrating the cup, I would suggest using the cup itself as a sink to kill the naughty particles. Append Peak SOP to your cup geometry and scale it up and use that as a sink.
Think you can use Delete SOP too but it doesn't take in a bounding object like Group SOP so may not be a good idea.
eitht.
- eetu
- Member
- 606 posts
- Joined: 5月 2007
- Offline
I've had problems like this, and it ended up being a problem of cup collision thickness compared to flip volume resolution.
I guess the collision volume needs to be thick enough to be watertight when sampled into the behind-the-scenes volume that runs the flip sim. The flip volume voxel size is 2x particle separation by default (IIRC), so the collision needs to be more than that, probably twice that should be safe.
There might have been other reasons as well, but in the end maiking the collision object thicker is the thing to do. Usually it doesn't matter if the glass collision object extends outside the glass itself.
I guess the collision volume needs to be thick enough to be watertight when sampled into the behind-the-scenes volume that runs the flip sim. The flip volume voxel size is 2x particle separation by default (IIRC), so the collision needs to be more than that, probably twice that should be safe.
There might have been other reasons as well, but in the end maiking the collision object thicker is the thing to do. Usually it doesn't matter if the glass collision object extends outside the glass itself.
- miklem
- Member
- 43 posts
- Joined: 4月 2012
- Offline
-
- Quick Links