Easy question on bubble level

[alive15]

Good afternoon guys, I have a basic question on the bubble level in the picture attached. This is probably really easy, but I don't get it, haha. There are two bubble levels here on the same axis. The bottom one shows the correct level, but the top one is always off? I don't understand why it's off, lol.

A co-worker said it's used when working upside down, but I don't know if he was being serious or pulling my leg! They got me good at work, please enlighten me so I may sleep in peace, haha.

*edit rotate photo; 2nd edit: it's 90° more ccw than it should be, my bad

[_Dock_]

The vials are slightly concave, or convex depending on how your looking at it. Since it has the concave, the level would not work if “upside down” by having two vials it is not directional. When upside down the bubble will find the highest point and doesn’t have the ability to over power the arc, which gives it the look of being way off.

[_Dock_]

Some grade levels that are convex go even further than that. If the bubble is touching either side of the line it will usually correlate to a certain percent in slope. Useful for a plumber determining fall on a drain line.

[drbitboy]

I will never get this done before there are ten other answers, but anyway.


He was almost definitely pulling your leg.



The glasses are slightly curved, but once calibrated the bottom glass is the one to use, as the bubble will be at the top of the curve, and if that top of the curve is midway between the marks i.e. the bubble edges equidistant from their respective edges, then the edges of the entire assembly are level.


The second glass is so you don't have to think about which way to use the assembly; you can use whichever glass is the bottom glass for the way you grab it.


The curve is very slight, so the bubble position is analog-ish, and no binary - [level => bubble in center] vs. [not level => bubble at one end] - states you might get with a straight glass (ignoring surface tension/friction?). Also, I suspect over time you might calibrate your eye to know how much out of level something is by judging how much the bubble is off-center.

[Ken Roach]

They're telling the truth; a level can have either a curved constant-diameter tube or a variable-diameter tube, so the bubble can seek the top.

That level has slightly arched tubes, with the lines inscribed on the top, so it can work while sitting on either side.

[Ron Beaufort]

just for kicks ...
.

[roxusa]

The Double bubble is so you can use either side as stated above but actually you check level when you start with device from both sides to insure accuracy of the level and that the level is not bent or damaged. At least that is what I was taught

[drbitboy]

Originally Posted by roxusa

The Double bubble is so you can use either side as stated above but actually you check level when you start with device from both sides to insure accuracy of the level and that the level is not bent or damaged. At least that is what I was taught

Does "both sides" mean swap it end for end (for a horizontal level test)? Because if that is the case, then even if the vials are not lined up with the long edges, if the long edges are straight you can still tell when something is level, because the bubble will be offset the same amount toward the same end of the level.

[roxusa]

end to end to see if the reading is the same and then flip it to use the other one
to see if its the same.

[drbitboy]

Originally Posted by roxusa

end to end to see if the reading is the same and then flip it to use the other one
to see if its the same.

Yeah, I do that even on a single-straight-vial level; if it's trivial to do a calibration of the device with every measurement, then why not?

[alive15]

Ahh okay, thanks everyone! Yeah, I did not notice the concavity of the tubes. Welp, time to wait for a new intern or co-op, I'll have to pass along the embarrassment to them, haha. Cheers,

[I_Automation]

Sounds like the crew that is going to tell you they don't have enough Ohms here and sent you to get them a bucket of Ohms.