Water flowing uphill

[davision]

I have a simple geometry object -- it's a plane which was modified slightly to create a gradual slope and a slight channel within it.

I'm using Phoenix to pour liquid (from a very low height) on top of this object but the liquid is behaving strangely. For the most part it's acts normal, but some "branches" of the liquid are flowing uphill and there are no wind forces or anything that would cause this.

Can you explain what's happening or what setting I can change to fix it? Incidentally, I'm using the same basic settings that are in that McCarthy video tutorial for liquid -- the one that shows a pipe pouring liquid into a Jammed simulator.

[Ivaylo Katev]

is the wetting enabled?

[davision]

Yes, wetting is enabled. I tried resimulating with wetting turned off, but there is still some water going uphill (although it looks better than before). It's a very small hill, but it still seems unnatural for the liquid to move up it.

[Ivaylo Katev]

not sure what can be the reason, can i see the scene?

[davision]

Okay, I emailed you.

[Ivaylo Katev]

you mean the liquid behind the source? i think is not unnatural, the liquid has a weak inertia from the falling and the surface slope is very low, that produces the upward movement.

[davision]

I will readily admit that you know more about fluid behavior than I do, but I don't agree that the liquid in this scene is behaving naturally. I sent you a new scene. I increased the severity of the slope in both directions. I also added a MassFX simluation of a sphere rolling along the surface to show the fall line of the surface. The water is flowing down one slope (in X direction) but UP the cross slope in the Y direction. If you had water dripping out of hose, it would not do that in real life.

I have another question -- the liquid seems to be disappearing as it moves away from the source. Is it evaporating? How can I control this and keep it from dissipating?

[Ivaylo Katev]

well, sometimes i also have the feeling that the water reaches too high levels when is falling over an inclined floor, perhaps would be good to make a real test
about the disappearing - no, the water is just increasing its speed and this produces the feeling that it disappears. similar situation can be seen when a stram is falling verticaly - first the crossection just goes thiner, later the stream is broken to separate droplets. even the most common problem is not the disappearing, but the producing of water "from nothing", when the advection is classic.

[davision]

As for the uphill issue -- I don't know what's really going on... but to me, it looks like the liquid is acting somewhat like a gas. Like each 'particle' of water is pushing against the other and it's expanding outward to fill a space. I know liquid does that naturally, but not in a direction opposite of gravity. It seems like the expansion-like behavior is overcoming the gravity forces of the incline.

[Ivaylo Katev]

hm, i have no max at home, but it can be the classic advection. as i wrote in the previous post, the calssic advection has the ability to create liquid from nothing, this can be the reason for the expansion. try with FT advection to check the difference (if is not FT). in general FT is more suitable for flowing liquids, the classic is suitable for settled liquids.

[davision]

I had been using Classic. When I tried FT, the simulation just freaked out. The water goes shooting off to one side.