Mass calibration
A balance reading is a force, not a mass, so air, temperature and level all enter. Mass calibration compares weights of known class with the balance, or weights with weights on a comparator.
Instruments and typical accuracy
| Instrument | Typical range | Typical accuracy | Where it is used |
|---|---|---|---|
| Microbalance | Up to about 5 g | Readability 0.1 to 1 µg | Pharma, filters, research |
| Analytical balance | Up to about 200 g | Readability 0.1 mg | Laboratory weighing |
| Precision balance | Up to several kg | Readability 1 mg to 0.1 g | Production, packaging |
| Mass comparator | To 1 kg to 50 kg | Down to micrograms | Weight calibration |
| Platform or floor scale | 10 kg to tonnes | 0.01 to 0.1 % of capacity | Logistics, trade |
| Reference weights (OIML E1 to M3) | 1 mg to 50 kg | See OIML R 111-1 tables | Calibrating balances |
Weights and balances are specified separately: a Class F1 weight does not make a Class F1 result. The accuracy figures are typical ranges to help you compare instruments. Always use the specification of the instrument in front of you.
OIML R 111-1 maximum permissible errors (± mg)
| Nominal mass | E1 | E2 | F1 | F2 | M1 |
|---|---|---|---|---|---|
| 1 mg | 0.003 | 0.006 | 0.020 | 0.06 | 0.20 |
| 10 mg | 0.003 | 0.008 | 0.025 | 0.08 | 0.25 |
| 100 mg | 0.005 | 0.016 | 0.05 | 0.16 | 0.5 |
| 1 g | 0.010 | 0.03 | 0.10 | 0.3 | 1.0 |
| 10 g | 0.020 | 0.06 | 0.20 | 0.6 | 2.0 |
| 100 g | 0.05 | 0.16 | 0.5 | 1.6 | 5.0 |
| 1 kg | 0.5 | 1.6 | 5.0 | 16 | 50 |
A selection of rows. The standard lists every nominal value and the classes M2 and M3. A weight is conforming only if its error plus the calibration uncertainty stays inside the limit.
Reference standards and traceability
- Weights. OIML R 111-1 defines classes E1, E2, F1, F2, M1, M2 and M3 with maximum permissible errors, shape, density and magnetic limits.
- Balances. OIML R 76 and NIST Handbook 44 define classes and tests for non-automatic weighing instruments.
- Traceability. Weights are traced to the kilogram, now defined from the Planck constant, through national and accredited labs.
- Quality system. ISO/IEC 17025 with ASTM E617 or OIML R 111 for the weight classes.
How it is calibrated
Typical procedure
- Let the weights reach laboratory temperature (hours for large weights) and handle them with gloves or tweezers.
- Level the balance, warm it up, and run its internal adjustment if it has one.
- Test repeatability: load a test weight 5 to 10 times and take the standard deviation.
- Test eccentricity with the weight in the centre and four corner positions.
- Test accuracy at several loads across the range, including the upper end, ascending and descending.
- Correct for air buoyancy when the density of the load differs from the weights, and report with the uncertainty.
Where it goes wrong
- Draughts and vibration. Air movement and tables that move make the last digits meaningless.
- Temperature differences. A warm weight appears lighter because of convection.
- Dirty or damaged weights. Fingerprints add milligrams, and a chip changes the mass.
- Magnetism and static. Steel and plastic loads can attract the pan.
- Ignoring buoyancy. Water at 100 g reads about 0.1 g low compared with steel weights.
What goes in the uncertainty budget
- Uncertainty of the reference weights, including drift.
- Repeatability and resolution of the balance.
- Eccentricity and linearity.
- Air buoyancy, from air density and sample density.
- Temperature, draughts, magnetism and static.
Combine the contributions as in the uncertainty budget calculator, then decide pass or fail with a guard band as explained on the calculators page.
Calculators
Everything runs in your browser. Nothing you type is sent anywhere.