Initial Fillup of complex shape...again...

[ben_hamburg]

I know this has been asked a couple of times already (including by myself), but I´m still not quite sure, how to tackle this setup most efficiently:

I have a container (organic shape) with one hole in it.
I want to fill up the container until just below the hole.
From a source I want to add more liquid to the container, until it overflows through the hole.

Right now I´m following the advice from the manual, but I´m not sure how long it should take for the liquid to settle this way.
I created a dummy Emitter object roughly covering the volume, that should be filled up, but because so much liquid is discharged in such a short amount of time, it rises up pretty high first, before it settles, right now I´ve simulated until frame 300 and it is only starting to settle down.
Are these usual times for the liquid to settle down?
Or is there another way to achieve this?

I also thought about just covering a smaller area, that would cover only some part right below the liquid surface plus the part where the emitter for the overflowing is positioned and then using the "initial fillup" option in the liquids tab.
I can see the grid beeing filled up immidiately, I´m just not sure the filling up with the extra liquid source and the overflowing would work then.

There is also another thing about Emitters I can´t seem to get right:

How do I emit liquid into just one direction?
For example: I have a cylinder and delete all faces but one cap.
If I untick "solid object" in its phoenix properties it will not emit no matter if I set the liquid source to "brush" or "inject".
If I tick "solid object" it only emits for one frame.
And if I don´t delete the other faces it emits into all directions of the cylinder.

So how do I get one continuos stream into one direction?

Last but not least:

I once more can´t work in real scale, so an object that would in reality be 10 cm now is 10 meters.
I know I can change the scene scale in the phoenix sim, but I´m not quite sure in which direction...10 or 0.1?

[ben_hamburg]

Another thing:

I noticed some unwanted behaviour with that fillup method: Because of the very high velocity/ pressure of the liquid after discharge, I get some "stuck" liquid sometimes, that gets in the walls of the container and then seems to be leaking more fluid from there...(I can see a thin film of liquid constantly dripping/pouring from where the liquid hits the wall).

I haven´t increased the steps per frame bexond 6 yet and my grid is also only about 2 mill cells right now, so this might go away with higher grid resolution / SPF Value.
Actually...Is it possible to change SPF values during sim?
So I could use a very high SPF at the start of the sim (until the liquid hits the wall) and then just reduce it afterwards, when the liquid isn´t moving as fast anymore?

[Ivaylo Katev]

can you send the scene? i think there is something that we miss.
about the liquid in one direction - the sources have a parameter called "polygon id" by default it is 0 that means all faces are emitting, set it to 1 and only the upper cap of the cylinder will emit.

[ben_hamburg]

Ah, I didn´t know how to use the polygon ID correctly...I assumed I had to enter the polygon ID that should be emitting, so I assigned ID 2 to those polygons, but that didn´t work, when I also assigned ID 2 in the PHX Source...

I did a simulation with 25 mill cells and 10 SPF overnight and the results look even worse, I´m sending the scene to support now...

[ben_hamburg]

I´ve sent the scene to support.
I did some more testing though and it seems I can actually use the "initial fillup" option from the liquid tab for my setup.
The camera is inside the object, so it never sees the fluid outside beeing filled by this option and filling the object further by a source also works.
I set the boundaries to jammed, not sure if that is necessary.

I also noticed a problem with my scene scale: the liquid source always seemed to behave strangely (like it wasn´t emitting in the right direction), but when I lowered the scene scale to 0.01, it started to look right again...

[Ivaylo Katev]

perhaps i can't understand the problem, but when i changed your source to be in brush mode and the discharge to be 100, everything worked fine.

[ben_hamburg]

Damnit. You´re right.
I must have at some point changed the source mode to "inject", and then of course I´m getting all sorts of problems...

Anyways, in my case the "initial fillup" method works even better, since I don´t need to calculate the 150-300 frames until the liquid settles and I can use a much smaller grid (at least I hope so, haven´t done a test render yet to see how the liquid looks like when some of it is cut off at the bottom...).

Thank you fpr looking into this though!

[ben_hamburg]

And damnit again...client wanted to change the camera animation and now i can´t use the initial fillup method anymore.
So I did a simulation overnight with the "brush mode 100-0 discharge method" and I am still getting some problems where liquid gets stuck between the walls of the container and starts to produce chaotic turbulences, but I guess that is a problem I can only tackle with higher SPFs.
Still haven´t tried to change the spf value in the middle of the sim though...(to only have high enough spf at the start of the sim when the initial fillup source makes those high velocity splashes, but not during the rest of the sim).

[tashko.zashev]

Hi Ben,

Try to simulate the first part, using higher SPF - you should get better results. It is possible to animate this parameter and get different SPF value for some parts of your timeline, but most probably this will lead to other issues. So it will be better to avoid doing it.