Resolution (measurement technology)
The resolution in the physical context is the limit of the ability of a device or a test arrangement to be able to separate values for a physical quantity from one another. The resolution therefore indicates the smallest perceptible difference. These can be voltages, angles, distances, frequencies or any other physical quantities.
The basic standard for measurement technology defines resolution as an indication of how far a measurement device can clearly differentiate between measured values that are close together; quantitatively, it can be specified by the smallest difference between two measured values that a measuring device clearly differentiates. The measurement deviation that occurs jointly with both values is irrelevant.
The resolution is generally smaller (“better”) than the systematic deviation . However, this can also be the other way round: If the resolution covers the systematic deviation with the appropriate accuracy , the measurement uncertainty can be reduced by repeating the measurement and averaging .
Small changes in measured values can be resolved better by means of a difference-forming measuring method , for example Wheatstone bridge , than when observing the measured variable itself.
Absolute and relative resolution
The resolution can be specified in two ways:
- absolute, i.e. in the unit of measurement of the size in question,
- or relative, i.e. as a numerical ratio. The denominator says
- when it comes to the characterization of the measuring device, the measuring range end value ,
- if it is a matter of an achieved or required resolution for a measured value that is already approximately known, this value itself.
Example: in the KATRIN neutrino experiment , electrons with kinetic energies of around 18.6 keV (kilo- electron volts) must be distinguished from one another to at least 1 eV (electron volts). The required absolute energy resolution is 1 eV, the relative energy resolution 1/18600 ≈ 5 · 10 −5 .
Resolution for the measured value display
In principle , a measured variable can be represented as finely as desired on a scale , but reading the measured value on the scale is only possible with limited resolution. The reading is limited to 1 ⁄ 2 … 1 ⁄ 10 scale division , depending on the design of the scale . With a scale divided into millimeters, about ½… ⅕ mm can be resolved without further aids. In the picture shown , lengths of 0.01 mm (½ scale division on the slide) can still be differentiated using the vernier scale. Ultimately, a measured value read on a scale can only be estimated in terms of its fineness; it is subject to an estimation uncertainty .
A measured value can always be read off exactly on a numeric display , but here the resolution for displaying the measured variable is limited, because only a number of different values can be displayed, and each display value stands for a partial range within the measuring range . The smallest difference between two measured values that distinguishes such a measuring device clearly arises usually from an increment ( digit ) on the least significant digit. For a voltage measuring device that divides the measuring range 0… 2000 mV into 2000 steps of equal width, the resolution is 1 mV. In addition to this absolute resolution, a relative resolution (one step related to all steps in the measuring range) can also be specified, in this specific case 1/2000.