Guide · Scales and weighing
Espresso scale accuracy vs resolution
A scale that shows 0.1 g is not necessarily accurate to 0.1 g. What resolution, repeatability and linearity mean, and how to check your scale at home.
We may earn a commission if you buy through links on this page, at no extra cost to you. Unless a page says otherwise, our picks come from published specifications and listing research, not hands-on testing. How we make money · How we research
Short answer
Resolution is the smallest step a scale displays; accuracy is how close the reading is to the true weight. A 0.1 g display does not promise 0.1 g accuracy. Look for a repeatability figure as well, and check any scale yourself with a known weight. For espresso, a scale that reads in 0.1 g steps and repeats itself is enough.
Four specs in plain words
| Spec | What it means | Example from a maker's manual |
|---|---|---|
| Resolution (readability) | The smallest step the display shows | Acaia Lunar: 0.1 g, or 0.01 g in its high-resolution mode[1] |
| Repeatability | The same object gives the same reading, again and again | Lunar: 0.1 g[1] |
| Linearity | How well accuracy holds across the weighing range | Lunar: 1 d, where d is 0.1 g[1] |
| Capacity and minimum weight | The heaviest load it takes and the smallest it registers | Lunar: 2,000 g maximum, 0.1 g minimum[1] |
Those are the maker's specifications, not an independent accuracy test. They are still useful, because they show what a complete spec sheet looks like. Compare that with the two scales in our catalog: their listings give a capacity (2 kg and 3 kg) and 0.1 g steps, both seller claims, and we have no repeatability or linearity figure for either.
Why resolution is not accuracy
Resolution is a property of the display. Accuracy is a property of the whole scale: the load cell, its calibration and how it handles temperature and time. A scale can show 0.1 g steps and still read a few tenths high if it is out of calibration, or drift while a cup sits on it.
Think of it this way: resolution tells you how finely the scale can talk; repeatability and linearity tell you whether it is telling the truth. A scale with a coarse display but steady readings is more useful than a twitchy one with an extra decimal place.
How much precision espresso needs
Makers' own examples give a sense of scale. Breville gives an 18 g dose for its 54 mm Bambino and Barista Express and 22 g for its 58 mm Dual Boiler.[2][3][4] Cafelat's Robot example turns 16 g of coffee into a 32 g drink.[5] Socratic Coffee's tamping experiment used 20 g in and 40 g out.[6]
At those sizes, 0.1 g is about half a percent of a 20 g dose and a quarter of a percent of a 40 g shot.[6] That is fine enough for home espresso. A 0.01 g mode is not needed for the dose or the shot.
The output weight matters as much as the dose. In Socratic Coffee's experiment, brew weight, not tamp force, drove the differences in strength (TDS).[6] That is the strongest practical reason to weigh the shot, not just the beans.
Speed matters too. If the reading lags behind the liquid, you stop the shot late and overshoot your target. Watch how quickly the number settles when you add a small weight.
Check your scale at home
- Put the scale on a firm, level surface away from drafts, and let it warm up briefly.
- Tare it empty. It should read zero and stay there.
- Weigh a known reference, ideally a calibration weight near your usual load. Note the reading.
- Repeat five times, lifting the weight off each time. The readings should match within one step. That is your home repeatability check.
- Check across the range. Weigh the reference alone, then on top of an empty cup after taring the cup. Both readings should match.
- Watch for drift. Leave a cup of water on the scale for a minute. The reading should not creep.
- Recalibrate if the manual describes how, then repeat the checks.
If a scale fails steps 4 or 6, no amount of resolution will save it.
Water, Bluetooth and safety marks
Water. Do not assume any coffee scale is waterproof. Acaia's Lunar manual gives no water-resistance or IP rating, lists error codes for water in the Power and Tare buttons, and says to clean it with a soft cloth and warm water only, without detergent or other chemicals.[1]
Bluetooth. In the US, devices that intentionally emit radio energy must be certified by the FCC or a certification body before they are imported or marketed, with some exemptions.[7][8] The rule does not name Bluetooth; our reading is that a Bluetooth scale is such a device. Either way, an FCC ID is not a food-safety or quality mark.
"UL approved". OSHA says its recognition of a testing laboratory such as UL is not an endorsement of the equipment that lab certifies.[9] A UL mark means a single testing laboratory certified the product, not that the government tested it.
Next
Questions
Do I need a 0.01 g scale for espresso?
No. Makers' example doses run from about 16 to 22 g, so a 0.1 g step is around half a percent of the dose. A scale that repeats its readings matters more than a finer display.[5][4]
What does repeatability mean on a scale spec?
How closely the scale gives the same reading for the same load, time after time. Acaia's Lunar manual, for example, lists 0.1 g repeatability alongside 0.1 g readability.[1]
Is my espresso scale waterproof?
Only if its maker gives a water-resistance (IP) rating. Acaia's Lunar manual gives none and lists error codes for water in its buttons, so do not assume a coffee scale is waterproof.[1]
Does an FCC ID mean a scale is safe?
No. US rules require radio transmitters to be certified before they are marketed, which we read as covering Bluetooth scales. An FCC ID is not a food-safety or quality mark.[7][8]
Why weigh the shot as well as the dose?
Because output weight changes strength. In Socratic Coffee's tamping experiment, brew weight, not tamp force, drove the differences in TDS.[6]