Behavior of bubble in a level

[HMS_Irruncible]

Can someone explain this observation… if I have a steel ball bearing on a glass plate, and I tip the plate even slightly to one side, the bearing will roll completely off. I’d expect similar behavior from a bubble in a level, but it always seems like it takes a heck of a lot more than a slight incline to drive the bubble to the end of the tube.

What is it about levels that makes this so? Is it the shape of the tube, the composition of the liquid, or just some quirk of fluid dynamics that makes the bubble so resistant to being displaced?

[RealityChuck]

The bearing rolls due to gravity, which isn’t pulling the bubble downward.

[beowulff]

The tubes are curved.
If they were straight, you would see the same behavior.

[retterath]

@beowulff

The tubes are curved.
If they were straight, you would see the same behavior.

That’s what I thought at first, but doesn’t your level work upside down too?

[beowulff]

@retterath

That’s what I thought at first, but doesn’t your level work upside down too?

From Wikipedia:

Where a spirit level must also be usable upside-down, the banana-shaped tube is replaced by a barrel-shaped tube. The upper internal surface of the tube is thus always of the appropriate shape.

[retterath]

Alright, now I have to go out to the garage to look. . .

[pericynthion]

The curvature is pretty subtle in a spirit level. It’s more pronounced on ball inclinometers such as the one found in an aircraft turn coördinator.

[Lumpy]

I’d always presumed it was surface tension making the bubble “sticky” relative to the liquid it’s in. I thought the whole point of “spirit” levels was using a liquid with a lower surface tension.

[Harmonious_Discord]

My dad’s old spirit levels have two parallel tubes in them aligned with the level’s two parallel surfaces, and you have to watch whichever tube is the top one. It also has two parallel tubes at 90 degrees to the level surfaces and two parallel tubes at 45% to the level surfaces. the total tubes in the level total six. Thus Wikipedia is not always right. They are all banana tubes in these levels.

[beowulff]

@Harmonious_Discord

My dad’s old spirit levels have two parallel tubes in them aligned with the level’s two parallel surfaces, and you have to watch whichever tube is the top one. It also has two parallel tubes at 90 degrees to the level surfaces and two parallel tubes at 45% to the level surfaces. the total tubes in the level total six. Thus Wikipedia is not always right. They are all banana tubes in these levels.

No, that entry is 100% right.
It doesn’t say anything precluding making a level with multiple curved tubes to allow use upside-down.

[jrepka]

Besides curvature, I see three possibilities:

1. density - the fluid has greater density than the gas in the bubble, is therefore harder to push out of the way;

2. viscosity - the greater viscosity of the fluid makes it more difficult for the bubble to push it out of the way;

3. surface tension - the fluid sticks to the plastic or glass tube better than the gas bubble does, and in order for the bubble to move through the tube it must detach the fluid ahead of it…

It may be none of these, one of these, or a combination of these, but the ball bearing certainly is not affected by any of these factors.