When a marking is worn away, two measurements can narrow the candidates. They cannot establish chemistry, voltage or device approval, so the result must remain a shortlist rather than a verdict.
Measure the two useful dimensions
Use a clean calliper. Measure diameter across the widest circular edge, then measure the total height from the flat negative face to the highest point of the positive cap. Do not squeeze the cell or include packaging. Record millimetres with the precision your tool genuinely supports.
A ruler may distinguish a 20 mm coin cell from a 16 mm one, but it is rarely precise enough to separate CR2032 from CR2025. Their 0.7 mm height difference is the entire physical distinction.
Choose a realistic tolerance
Published values are nominal and handheld measurements are imperfect. Start with ±0.2 mm for a decent calliper. Increase tolerance only when the measurement is uncertain, and inspect the distance of every candidate rather than assuming the first result is correct.
Measure twice at different rotations.
Use a decimal point, not fractions of an inch.
Reject zero, negative or obviously impossible values.
Sort candidates by combined diameter and height difference.
Finish the identification
After finding candidates, compare the surviving markings, nominal voltage, chemistry and drain category. A 7.9 × 3.6 mm measurement can describe alkaline LR41, silver-oxide SR41 and zinc-air PR41 families. Those are not one electrical product.
If the device documentation names a code, that requirement is stronger evidence than a physical measurement.
A practical calliper sequence
Close the clean jaws gently around the widest part of the rim and note the diameter. Rotate the cell and repeat; a tilted reading is usually too large. Then place the flat faces between the jaws to measure total height. Record both readings before searching.
If repeat readings differ by more than the selected tolerance, stop and inspect the tool position. Increasing tolerance until a desired code appears defeats the purpose of measurement.
How the candidate list should be read
The closest result is a geometric neighbour, not a recommendation. Compare the absolute diameter difference and height difference separately. A candidate with a perfect diameter but a 0.7 mm height gap may be less useful than the label resemblance suggests.
For example, a 20.0 × 2.5 mm measurement points toward CR2025. CR2032 shares the diameter and electrical family but its 3.2 mm height remains outside a careful ±0.2 mm search.
Limits of measurement
Corrosion, swelling, damaged wrappers and attached tabs can distort a reading. Do not handle a leaking, hot or visibly damaged cell merely to identify it. Measurement also cannot reveal nominal voltage, chemistry or drain designation.
Where the old cell is missing, measure the compartment only as supporting evidence and rely on the device label or manual for the required code. Compartment clearances are not the same as cell dimensions.
Measurement checklist
Before accepting a candidate, confirm that the calliper was zeroed, the cell was measured at the widest diameter, and height was taken across the complete can without squeezing. Repeat each measurement. Enter the observed decimals and an honest tolerance, then inspect the dimensional difference for every candidate. Finish by checking the printed code or device requirement.
Measure in millimetres.
Repeat at another rotation.
Treat corrosion or swelling as a stop condition.
Never use the candidate list as a replacement verdict.