Foam Particles too Large/Some Areas Not Filling In with Water

[hrigsby]

I'm working on my first non-tutorial effort in Phoenix and I figured I'd check in and try to get some general feedback. I've posted about it before, but I'm making a river with a waterfall portion. I've done this with two cascading Phoenix Liquid sims. I then resimmed foam, splashes, and mist on top of the original liquid sim. The bubbles for this sim are way too large. I assume I need to adjust the particle size (can I do this for just the foam and not the other particles)?

I also can't get the water to fill in some areas around the rocks. I assume this relates to the particle size of the water not reacting to the surfaces the same way an actual water particle would? Is there any way to adjust this other than to reduce the particle size and resim?

Any other feedback? Thanks in advance for any help! I think I'm getting closer to understanding the parameters, but there's a lot to experiment with.

[Gosho.Genchev]

I assume I need to adjust the particle size (can I do this for just the foam and not the other particles)?

Yup, that's possible. I'm going to assume you are using 3ds Max... So, each particle type (Foam, Splashes, Mist, Drag, Liquid) are stored in a separate Particle Group which you can access from the Phoenix Particle Shader. Create a Particle Shader, click the Add button at the top and select the Phoenix FD Simulator. A window will show up asking which particle group you want to select. At this point this new Particle Shader will be used for shading those particles.
To be fair, I would save my scene in its current state, make a new version, remove all Particle Shaders currently in the scene and create my own. Name them appropriately and tweak them as you see fit.

I assume this relates to the particle size of the water not reacting to the surfaces the same way an actual water particle would?

Lowering the Grid roll-out -> Cell Size will help in case the simulation is simply too low-res for the amount of detail in your geometry.
If the problem is that the water is moving too quickly, increasing Dynamics -> Steps per Frame will resolve the issue.

By the way, we have a River tutorial in case you'd like to take a look. You can find it here: https://docs.chaosgroup.com/display/PHX3MAX/River
It's nothing fancy but there are some good tips related to wetmaps and speeding up your simulation with Initial Fill geometry.

Cheers!

[hrigsby]

Yup, that's possible. I'm going to assume you are using 3ds Max... So, each particle type (Foam, Splashes, Mist, Drag, Liquid) are stored in a separate Particle Group which you can access from the Phoenix Particle Shader. Create a Particle Shader, click the Add button at the top and select the Phoenix FD Simulator. A window will show up asking which particle group you want to select. At this point this new Particle Shader will be used for shading those particles.
To be fair, I would save my scene in its current state, make a new version, remove all Particle Shaders currently in the scene and create my own. Name them appropriately and tweak them as you see fit.

Cheers!

Ah, I see now. It looks like I can't adjust these parameters (primarily "Size Multiplier" and "Count Multiplier" without resimming? It seems like it would be possible to do without resimming given the description of what these parameters do, but y'all know better than I do!

[hrigsby]

I changed the "size multiplier" to .1 and the "count multiplier" to 10 for both the foam and splash shaders. Then I upped the liquid sim "steps per frame" to 4 (from 1) and now my splash/foam/mist particles are flying all over the place at the slightest collision. What did I do wrong here?

[Gosho.Genchev]

Ah, I see now. It looks like I can't adjust these parameters (primarily "Size Multiplier" and "Count Multiplier" without resimming? It seems like it would be possible to do without resimming given the description of what these parameters do, but y'all know better than I do!

Oh, that's totally possible. You can adjust any parameter in the Particle Shader and hit render to see the update - there's no need to start a new simulation in that case.

The Particle Shader is basically a material for the selected particle group. Its parameters do not affect the simulation in any way whatsoever.


Regarding the exploding particles - could you please upload a screenshot of your simulator settings (and please! make sure to open all roll-outs) so we can check what the problem may be ?

[hrigsby]

Sure thing. I also realized that I upped the Liquid Source Velocity from 2' to 2' 6", but it doesn't seem like that should have created such a gravity defying sim (the water particles started going crazy too after a few more frames). I've since reverted the velocity back to 2' so I'll see if that helps.

I've attached the rest of the settings. Thanks again for your help.

[Gosho.Genchev]

the water particles started going crazy too after a few more frames

Oh, in that case it might be caused by the simulation interacting with bad geometry - holes, overlapping faces, etc could give you an exploding simulation. You could try applying an STL modifier to your geometry to check for such issues: https://knowledge.autodesk.com/suppo...68C7F-htm.html

Is the ground geometry in you scene a thin sheet (e.g. a plane)? Can you please try applying a Shell modifier to see if that resolves the problem ?

Looking at the screenshot, your settings seem reasonable to me. In general, a quick and easy way to check for issues is to:

1. Hide all geometry except the ground
2. Disable Foam and Splash from their respective roll-outs
3. Increase Cell Size until Total Cells is about 5 or so million
4. Run the simulation for 50 or so frames - should only take a few minutes.
5. If it runs OK, enable Foam, then Splash, then reduce Cell Size a bit, etc.


If none of the suggestions above help, please send the scene over to support@chaosgroup.com so we can take a look and help you out.

[hrigsby]

The terrain is shelled. For some reason, reducing the velocity back to 2' seems to work. Is there any rhyme or reason to the velocity being a measure of distance? Is that like "feet per second" or something similar to that? It doesn't seem like such a small change (6") should make it go crazy like that. I could see the water hitting the terrain geometry slightly faster making some splashes, but flying 20' in the air vs. moving normally along the ground is a bit extreme and makes it seem like something odd is going on (as you were getting at).

I'll mess with the STL modifier once the sim finishes.

[Svetlin.Nikolov]

Hey,

The velocity is a distance since Max or Maya do not support velocities or accelerations - it should be read "per second" One more thing that bothers me is surface tension - for such a large scale sim you don't need it at all and it might make the liquid bulky if enabled.

How does the source work? Does it emit from the bottom as well, or only from an emitter to the side? One more thing - in the previous thread, what was the reason that liquid did not cover some parts of the river bed? Now I'm thinking it could be related to the surface tension, and also it could be related to open geometry.

One final note - foam sizes are controllable at simulation time by the Foam's Size option, while Splash sizes during simulation depend on the Grid resolution. The larger your voxels, the larger the Splashes. You can reduce these after simulation at render time as Gosho said, using the Particle Shader, but note that if two particles were pushing against each other during simulation, if you reduce their render sizes, a gap will appear between them, so keep this in mind.

Cheers!

[hrigsby]

Hey,

The velocity is a distance since Max or Maya do not support velocities or accelerations - it should be read "per second" One more thing that bothers me is surface tension - for such a large scale sim you don't need it at all and it might make the liquid bulky if enabled.

How does the source work? Does it emit from the bottom as well, or only from an emitter to the side? One more thing - in the previous thread, what was the reason that liquid did not cover some parts of the river bed? Now I'm thinking it could be related to the surface tension, and also it could be related to open geometry.

One final note - foam sizes are controllable at simulation time by the Foam's Size option, while Splash sizes during simulation depend on the Grid resolution. The larger your voxels, the larger the Splashes. You can reduce these after simulation at render time as Gosho said, using the Particle Shader, but note that if two particles were pushing against each other during simulation, if you reduce their render sizes, a gap will appear between them, so keep this in mind.

Cheers!

The primary source is the rectangles behind the bridge. They just omit particles out of the bottom (face ID) and then I added the spheres, which emit over their entire surface, to try and fill in some areas that weren't getting enough liquid. None of the actual river geometry (rocks, riverbed, etc.) emits any liquid. I struggled trying to get the correct volume of water flowing down the river to make it so that it doesn't overflow the banks but also doesn't leave areas unfilled. I'm sure there are many better ways to achieve what I was going for.

I actually added the surface tension in an attempt to keep rogue water particles from breaking out of the main water body and climbing up the river banks. I've since cut it off per your instructions.

In the last thread, I had made a newbie mistake and part of the riverbed was below the grid (which was set to open so that the cascade sim would work); the water that hit the bottom of the grid was disappearing and once I lowered/expanded the grid that problem was fixed. There could definitely still be other issues, but I'm certain that the terrain is shelled and doesn't have any holes in it. Could the rocks intersecting and being buried in the terrain be causing a problem?

[Svetlin.Nikolov]

Hmm, the rocks setup does not sound worrying so maybe it's something else. If you can, send the scene over to support so we can check out what causes the water not to fill the area around rocks. Did you figure what is causing the particles to go wild or should we see what causes this?