Shaping a flip fluid pour?

[nigelgardiner]

Hi there

I'm trying to determine an approach to shaping an emitted flip fluid pour.

When you look at fluid poured from a bottle or a jug there's an awful lot going on in the way the fluid surface behaves. it very nicely rolls i guess (not the best description) into itself from the sides and tapers in on itself the further away it is from the top of the pour. It's a really beautiful shape.

I should imagine the physics of it is pretty complicated but does anyone have any ideas about a basic plan of attack or know of any documentation that discusses this?  I've had a hunt around but don't seem to be able to find anything.

I guess I want to be looking at curvature and/or surface tension or something like that and manipulating the velocities on the particles at the edges of the pour relative to those at the centre?

Any advice much appreciated, this is just a learning exercise at this stage.

thanks

Nigel.

[Guest tar]

The 'swirly kernel' is meant to help with all that. If you are comfortable with the mathematics then you want to read this:

'FLUID SIMULATION SIGGRAPH 2007 Course Notes'

https://www.cs.ubc.ca/~rbridson/fluidsimulation/fluids_notes.pdf

[nigelgardiner]

Hi Marty

Thanks, I'm already using the swirly kernel.

And ta for the fluid notes, much appreciated.

Have a good day.

[Guest tar]

I'm not exactly sure the effect you are after but surface tension has the largest effect on the surface, much more than air e.g Does water temperature/impurities affect this look you are after in real life? more of both lessen the surface tension.

[Atom]

The simplest way is just to model the "spout" and set it as a collision object. Then the fluid will be forced into the shape of the pouring vessel.

 

Edited August 16, 2017 by Atom

[nigelgardiner]

I want to emulate the flow of liquid from the neck of a bottle as per the attached image and out of interest I'm trying to determine how I could control the way it shapes.

Sorry I don't know what you call whatever is happening in terms of physics but the liquid in that shot (and real life pours in general) has such a nice shape and taper to it. To date I haven't been able to achieve anything close.

I did try and cheat it with noise on the velocities from the emitter volume but that looked terrible.

Quickly looking at the fluid notes Marty sent thru earlier, there seems to be a few relationships that could be useful but it's going to take a close read for me to understand it, if i even can.

 

cheers

Nigel.

pour.tiff

[ryew]

Before you invest too much time in custom velocities fields, I would definitely second Atom's suggestion of adding some subtle ribs or other cheat geometry detail to your collision object for starters as that will help influence the way velocities evolve in the sim over time. 

It's also always a good idea to post an example scene demonstrating your current attempts as well if it's not in violation of any confidentiality agreements

[nigelgardiner]

Hi Ryew

Yep pouring a volume of fluid through a spout is definitely an improvement on emitting from a source volume.

I'd tried emitting from a source volume thru a spout before and it didn't really work but pouring a bottle or glass of fluid starts to give a better shape.

The paper Marty attached above explains the Navier Stokes equations pretty well. 

Using a spout defines a solid boundary for the time the fluid is moving thru it but then it's defined entirely by a free surface when traveling thru air, surface tension is the major factor here, (thanks Marty) and I think that's able to be defined by mean curvature which you can solve for with gas analysis micro solver, and then, well I'm still figuring that out.

I read the Flip smorganic/sheeter effect thread last, Pazuzu's tension work was amazing.

Haven't got a flip worth sharing at the moment but will do.