Underwater Bubbles!

[GrahamMac]

In Phx 2.0 I have a wave tank setup and am trying to generate bubbles rising into the shot. It is important that the bubbles are not simple spheres (i.e. foam particles) as they must be seen to be buffeted and distorted by the surrounding liquid.
At the moment I am doing this with a spinning paddle (out of shot) which creates the bubbles but also generates a lot of turbulence at the surface of the liquid.

Is there a more efficient way of creating extra bubbles in liquid?

Thanks

[Ivaylo Katev]

you can use PF to drive the uw bubbles and to render them with an air material (low ior, let say 0.1)
here is the test:

[wxyz]

Can you explain this technique in more detail?

[Ivaylo Katev]

create the bubble geometry, perhaps the best way is to apply some modifiers over a simple sphere. create the PF system, and replace the default shape operator by a shape instance one, and select the bubble geometry. switch on the "animate shape" option, and animate the modifiers to make the bubbles to change their form. assign the air material to the geometry and this is all.

[Paul Oblomov]

Not quite undersand why you can't use pf as a particle system, that phx foam understand ?

[Ivaylo Katev]

Not quite undersand why you can't use pf as a particle system, that phx foam understand ?

he need not spherical bubbles, phxfoam produces spherical bubbles

[GrahamMac]

Thanks Ivaylo.
I will experiment with the PF approach. The key will be in making the bubble shape animate convincingly with the motion of the particle.

[Ivaylo Katev]

one more thing - i wrote that air's ior is less than 1, this is wrong, it is 1
the result is not very different but just for accuracy....

[GrahamMac]

I had another idea of how to create bubbles in liquid. Using a phoenix source emitting fluid at temperature 0 from within the liquid.
Before trying the PF approach I wanted to see if this might work so I set it up and am getting bubbles without adding any turbulence to the liquid.

What I’m finding though is that these bubbles tend not to want to join together or become slightly flattened as they rise through the surrounding liquid.
Instead they tend to:
1) Form long vertical bubbles that stretch thin and occasionally break apart into multiple smaller bubbles.
2) Many smaller bubbles also seem to shrink and disappear before getting to the surface. (without turning into foam)

To solve (1)
I though some surface tension (with action distance 4) might help reduce the verticality, but it doesn’t appear to help much.

To solve (2), cooling was already set to 0 thus preventing liquid cooling into air (temperature 1 to 0) so I thought this should stop the opposite (air to liquid – bubble shrinks).

For both issues I’m not sure what else can be done to improve the simulation.

Any and all ideas appreciated!

[Ivaylo Katev]

yes, the surface tension plays an important role in the natural bubbles, but your second suggestion is wrong - actually the conservation is to blame. the liquid is about 1000 times denser than the air, and because of this it overcomes its pressure resistance easily. in the real world, there is a moment when the air pressure is strong enough to stop any further compression, but such a high pressure requires very high conservation levels to be simulated. the simplified rule is you need conservation value above the density ratio, that is 1000 in our case.

[GrahamMac]

Thank you Ivaylo,

It took a fair bit of playing around with the settings but I am getting better looking bubbles now with higher surface tension and conservation quality in the region of 200 to 1000.