NavList:

Noell Wilson,

You wrote to me: “I agree the accuracy of a level is specified movement per foot – or some other distance per distance.”

In my last post to Marty Lyons I made this statement about level sensitivity, not accuracy: “Technically, spirit level sensitivity is the angular change in inclination that causes the bubble to move a specified distance.”

Two examples of how sensitivity can be unambiguously expressed are: 20 arc seconds / 2 mm and 0.5 degrees / 0.1 inch.  Manufacturers who advertise the sensitivity of their level products often leave off the bubble movement distance altogether, or substitute “per division“, leaving it to the potential buyer to dig deeper to find what the movement distance is. Worse (in my opinion), is the tendency to specify a slope (rise / run) alone, rather than an angle and bubble movement distance; for example, 0.5 mm / m,  0.05 inches / 10 inches. 

Sensitivity specification ambiguity could be avoided by simply specifying the radius of curvature of the level vial used, as it’s what determines how sensitive the bubble movement is to a change in inclination. A user could then calculate sensitivity by dividing a bubble movement distance of his choice by the circumference of the circle defined by the radius of curvature, multiplied by 360 degrees (or 360 x 60 for arc minutes or 360 x 3600 for arc seconds).

I’ve never found an explicit definition for spirit level accuracy. What follows provides a clue as to what two manufacturers take it to mean--for what I’d categorize as ‘carpenter’s levels’. These typically have vials with a radius of curvature of, or close to, 200 mm. (Note that they are using for accuracy the same sorts of specification format as I cited above for sensitivity when no bubble movement is specified, making it easy to confuse poorly written sensitivity statements with those for accuracy.)

Johnson Levels https://www.johnsonlevel.com/News/SpiritLevelInformation

“The best spirit level is accurate to within plus or minus 0.5 millimeters/meter, or 0.005 inches/inch or .029 degrees.”

Stabila 70T level  https://www.stabila.com/en/products/details/type-70-t-spirit-level.html

“Measurement accuracy in normal position   0.50 mm/m”

Both of the above statements are claiming a measurement accuracy of +/- 1.72 arc minutes.

I have a Stabila 70T level (image, with my note in red, attached). In 2017 I used a home-made level checker to measure the sensitivity / radius of curvature of it and several others (second attachment). The 70T had a radius of curvature of 200 mm. For a bubble movement of 2 mm that gives a sensitivity of 34.4 arc minutes / 2 mm. The fact that the angle in the claimed accuracy is exactly 20 times smaller than the angle in the sensitivity, suggests to me that they assume a perfect installation of the vial in the level body and that the bubble can be centered to within 2 mm / 20 = 0.1 mm.

The 0.1 mm centering assumption may be pushing the advertising envelope a bit as it’s five times smaller than the excerpt quoted below from this page:

https://www.leveldevelopments.com/2020/10/sensitivity-accuracy-of-spirit-level-vials/

“The accuracy at which users can centre the bubble in the vial should also be considered, which is usually assumed to be 0.5 mm. For example, if a product needs to be level within 0.1° (or 6’), a vial that will give at least 0.5 mm bubble movement for a 0.1° (6’) change in angle will be required. This corresponds to 0.4° (24’) for a 2 mm bubble movement, therefore a vial with a sensitivity of 24’/2 mm (24 minutes per 2 mm bubble movement) is needed.”

Chuck V.

File:

File: