Checking a kitchen scale with coins of published mass: a repeatability, range and corner-load record

Goal

Produce a dated sheet for one kitchen scale that shows how its readings relate to known masses at several points of its range, how repeatable a reading is, and whether the position on the platform changes it, so that later logs weighed on this scale carry a stated basis.

Prerequisites

A scale with its display resolution noted (typically 1 g or 0.1 g). Reference masses: new, clean coins whose nominal mass the issuing mint publishes. The ECB lists the euro coins, for example 7.50 g for 1 euro and 8.50 g for 2 euro; the Royal Mint lists the one pound coin at 8.75 g. The published figure is the nominal mass of a new coin; wear and dirt change it by an amount this protocol does not know, so use uncirculated or near-new coins and record which. A flat, level surface away from draughts.

Steps

1. Header: scale model, display resolution, stated capacity, battery state, surface, room temperature, date, coin type and count used, and the nominal mass per coin with its source.
2. Switch on with nothing on the platform, wait for a stable zero, and record it.
3. Single coin: place one coin in the centre, record the reading, remove it, wait for zero. Repeat five times. Record all five readings, not an average.
4. Range: build stacks of 5, 10, 20 and, if the coins allow, 50 coins; record the nominal sum (arithmetic from the published value) and the reading for each, placed once in the centre.
5. Corners: with the 20-coin stack, record the reading at the centre and at each of the four corners of the platform, about 2 cm from the edge.
6. Hold: leave the 20-coin stack in the centre for 60 s and record the reading at 0, 30 and 60 s.
7. Tare check: tare with a bowl, add the 10-coin stack, record.
8. Write the differences (reading minus nominal) in a separate column; keep the raw readings.

Expected result

A sheet showing, for this scale on this date, the spread of repeated single-coin readings, the difference from nominal at each stack size, any corner-dependence and any drift during a hold. Repeating the sheet monthly shows whether these numbers change.

Limits and test basis

Proposed protocol; no reading, tolerance or verdict is claimed. Coins are not calibrated weights: their nominal mass is a minting specification, not a measurement of the coin in hand. A reading that differs from nominal by less than the display resolution cannot be distinguished from rounding. The protocol records; it does not adjust the scale or decide whether it is fit for any use.

Matching the test mass to the display resolution

A reading can only show what the display can resolve. On a scale with a 1 g resolution, a single 7.50 g coin reads 7 or 8 g and five repetitions show rounding, not repeatability; the repeatability and corner tests should then use the 20-coin stack (a nominal 150 g for 1-euro coins), and the single-coin repetition is reserved for scales with a 0.1 g resolution, where the same coin is 75 display units. Choose every test load so that the smallest difference worth recording is at least two display units, and write the resolution next to each test in the header so a later reader knows which differences the sheet could have shown.