Problem with molotow-cocktail - liquid through wall

[Sascha Selent]

Hi everyone.

I have some problems with making a molotow-cocktail. This is what happens:

Okay, so far it works fine to fill the bottle with liquid and the travelling through the air is also fine, but when it comes to impact, the liquid goes through the wall and partly sticks inside.

I think it has to do with timing - see that I move the unfractured bottle the frame before impact away from the scene - this works fine, but it seems that the liquid then has too less time to react to the wall.

So, what would you say is my problem and what would be the solution? Is the setup approach okay, or would you do it different?

I also attached screenshots of the 3 impact frames, so that you can see in detail what is happening. The max-scene is quite big, but if needed I can stripe it down for download.

And a question between: Is it true, that the standard gravity has it´s influence in the negative z-axis-direction of the simulator?

Thanks in advance.

Sascha

[Ivaylo Katev]

from simulator point of view the wall is moving and it's normal that some part of the liquid penetrates inside, actually the wall just overlaps some cells that contain liquid in the previous frame. in the rendering this liquid should be hidden by the wall and no visual problem should appear. of course if the wall is not thick enough, some liquid may appear behind the wall, that may be a problem. anyway, if you need to remove this side effect, the best way that i know is to make second instance of the wall, put it in the exclude list, make it brush mode air source (liquid=0). in the next tutorial series we will dedicate one to this problem, how to remove the liquid penetration in the moving geometry.

[Sascha Selent]

YOU ARE SO FAST! Absolutely superb!

Ivaylo!

Thank you for your quick answer!

if you need to remove this side effect

Yes, I do, because the wall is only a placeholder. Finally there will be a thin sheet of glass, so eveything behind will be visible...

the best way that i know is to make second instance of the wall, put it in the exclude list, make it brush mode air source (liquid=0).

Okay, can you explain this further?

Instance of the wall, okay, no problem.

But in which exclude list do I have to put it? From the source?

And what exactly do you mean with "brush mode air source (liquid=0)"? Do I have to set this in the source only? Do I need another source?

Again, thaaank you for your fast support! I do appreciate this...

Sascha

[Ivaylo Katev]

fortunately yesterday we had a question form an user with similar problem, he have a moving grove with flowing liquid, and the liquid goes trough the grove. here is the scene with the solution, your situation is exactly the same

[Sascha Selent]

Wooaaaahhh! Grrreat!

Ivaylo, can´t say it often enough - your support is absolutely amazing!

Eeeehm, is it Max 2012?

[Ivaylo Katev]

yes, 2012. the lack of forward compatibility is really annoying, the most annoying "feature" of 3dmax and the most funny part is that max uses chunk concept to write its files, that gives good backward and forward compatibility, so the forward one is rejected with special efforts, by marketing reasons

[Sascha Selent]

yes, 2012. the lack of backward compatibility is really annoying, the most annoying "feature" of 3dmax

But you can save a scene in formats down to R 2010 (). Does that have impact to PFD?

Right now I´m adopting your settings to my scene, I´ll keep you up to date...

[Ivaylo Katev]

the export in 2010 format is correct, no damages are observed

[Sascha Selent]

Right now I have some weird problems with simulating the fuel.

I need a really high setting for the SPF upper limit - the SPF reaches this value at some moment and the fuel dissappears. When I go back some frames and do a restore it works again until the next reach of the SPF upper limit. The sim is not constant, it flickers.

I also experienced, that after a certain time of working the sim is significantly slower - after restart (not max, the machine) it is much faster again - but my RAM is not exceeded...

What can this be?

[Ivaylo Katev]

and how big is your spf upper limit?

[Sascha Selent]

30 at the moment...

...it appears, when the bottle is moving and rotating fast.

Maybe using Fuel is problematic? I would like to use the burning feature (does work with Fuel only, right?).

[Ivaylo Katev]

the spf is not caused bu the using fuel, it just needs to be high to keep the desired maximal step. but we are planning to abandon the ability to use any channel as liquid in the future and for simultaneous liquid and burning simulation. it seems this feature makes phoenix very complex for the most users, and the most important it makes the simulation a bit slower, and is a major reason for bugs.
btw, when the spf upper limit is reached, the maximal step limits the fluid transfer, that decreases the simulation quality and seems to be a big problem for the less experienced users, and even for the experienced ones. i would recommend you to set fixed spf rate (for example 2) and big advection step, but this makes the adaptive grid unusable. actually this is an option, switch off the adaptive grid and use 100/2/2 advection settings.

[Sascha Selent]

Ivaylo.

I´m trying to reproduce your setup, but I fail. May you have a look on my testscene? I must have missed a step...

[Ivaylo Katev]

in your scene the source helper is put in the exclude list, but it shouldn't be. the wall_eraser object should be instead

[Sascha Selent]

Man, thank you, yeah, that was a mistake, but it still does not work...

...did it simulate correctly with your scene?

EDIT: seems to be a scene-scale-problem. Your testscene works fine, but the objects are tiny. If I take this scene and bring the objects to my dimensions the liquid penetrates the wall again..