Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Help  Consistency in emitting fluid
#1
Triste 
Hi I've been trying to make a chocolate fountain with obi fluid but there're some problems I encountered.

1) The particles remain to be very big.
If I turn resolution up higher, a tiny portion of them will burst out with the remaining sticking to the emitter and won't flow down.

2) I want it to be as thick as ketchup but turning up viscosity has no effect.

3) They're bursting out once in a while but i want them to be sticky and flowing continuously from the emitter instead.

4) I tried to change the colour of the particles to be opaque but looks like there's no place for me to place the opaque material?

What can I do on the settings or is there scripts required?

Attached a video of how the particles look like now.
Reply
#2
(25-11-2021, 06:47 AM)inabird52 Wrote: Attached a video of how the particles look like now.

Hi,

Can't see any video link in your post, so answering a bit blindly Sonrojado :

(25-11-2021, 06:47 AM)inabird52 Wrote: 1) The particles remain to be very big.
If I turn resolution up higher, a tiny portion of them will burst out with the remaining sticking to the emitter and won't flow down.

Changing resolution is not the only way to resize your simulation (though it shouldn't cause any issues). You can just scale the solver down, see the video on this subject:

https://www.youtube.com/watch?v=W0

(25-11-2021, 06:47 AM)inabird52 Wrote: 2) I want it to be as thick as ketchup but turning up viscosity has no effect.

Make sure you're modifying the blueprint being used by your emitter. Also, keep in mind that viscosity (and a few other parameters) have a certain stability range:
http://obi.virtualmethodstudio.com/manua...rials.html

Quote:Viscosity:As a rule of thumb, this value should always be between 0 and 1.5-2.

Too high viscosity values can lead to simulation instability, depending on your timestep length. Shorter timesteps yield more stable simulations.

You can't make a fluid arbitrarily viscous while keeping the simulation timestep constant, as you go higher in viscosity you need to reduce the timestep which makes the simulation costlier. This happens in all fluid solvers, both realtime and offline. As an example, RealFlow takes around 10 hours to simulate a few seconds worth of toothpaste-like fluid: https://www.youtube.com/watch?v=oa43MoGSrRI

(25-11-2021, 06:47 AM)inabird52 Wrote: 3) They're bursting out once in a while but i want them to be sticky and flowing continuously from the emitter instead.

If they burst sometimes, probably you're very close to or outside the stability range for viscosity. Either reduce the viscosity or reduce your timestep.

(25-11-2021, 06:47 AM)inabird52 Wrote: 4) I tried to change the colour of the particles to be opaque but looks like there's no place for me to place the opaque material?

You can't use arbitrary materials on fluid, since fluids are not meshes. There's a dedicated fluid renderer in Obi that allows you to render opaque, transparent and flat materials (or anything in the middle), check out the manual for fluid rendering:
http://obi.virtualmethodstudio.com/manua...ering.html

Looks like the included FluidRaclette sample scene does something close to what you need in terms of "thick, opaque fluid", you might want to check it out.


let me know if you need further help. cheers!
Reply
#3
Thank you for replying Josemendez!

Sorry I didn't realise the video wasn't attached.
You can see the video by clicking this link.

It's much better after editing the blueprint itself now.
The particles look more like liquid but they're still splashing too much. They're not flowing along the surface of each dome but instead flying outwards.
If I decrease the buoyancy the fluid won't flow along the surfaces.

Here is the setting of the blueprint:
Capacity: 1500
Resolution: 1
Rest Density: 1000
Smoothing: 2
Viscosity: 0.15
Surface Tension: 1
Buoyancy: -20
Atmospheric Drag: 0
Atmospheric Pressure: 20
Diffusion: 0

Setting of the Obi Emitter:
Collision material: HighFriction (Obi Collision Material)
Phase: 1
Emission method: STREAM
Min pool size: 0.115
Speed: 0.5
Lifespan: 5
Random velocity: 0
Use shape color V

Obi Solver:
Backend: Burst
Mode: Mode 3D
Interpolation: None
Gravity: X0 Y-6 Z0
Damping: 0
Shock Propagation: 0
Max Anisotropy: 3
Max Depenetration: 5
Sleep Treshold: 0

World Linear Inertia Scale: 0
World Angular Inertia Scale: 0
Simulate When Invisible V

And nothing is changed in the Constraints.

Is there any changes I can make to turn it into something like thick liquified chocolate?
Reply
#4
Your fluid settings will result in a extremely dense gas, nothing like chocolate at all:

Viscosity: 0.15 //<--- way too low for any viscous fluid. Try setting it to 1.
Buoyancy: -20 //<--- will turn your fluid into a -20 times denser than air- gas. Set to -1.
Atmospheric Pressure: 20 //<--- will cause your fluid to collapse into thick drops, due to the huge air pressure. Set to 0.

Also, you're using the included "high friction" collision material, which as the name implies has high friction but zero sticking. This will make your fluid slide right off any surfaces.

You can use the included "ViscousFluid" sample scene as reference for most if not all of your fluid/collision settings.
Reply