Verification:- Verification of the balance is performed by placing a suitable weight (depending on the type of balance) in the centre of the weighing pan once and comparing the result with pre-defined acceptance criteria. The same weight should always be used in these verifications.

The acceptance criteria shall be defined by each individual OMCL.

Accuracy:- The accuracy of the balance is checked by weighing at least three different certified weights that cover the usual weighing range of the balance. It is recommended that the weights have approximately 5%, 50% and 100% of the maximum capacity of the balance (or of the maximum weight used on the balance), depending on the type of balance. It is recommended that the weighing is repeated at least 5 times for every weight, particularly, when the results shall also be used in the test for precision.

The acceptance criteria shall be defined by each individual OMCL.

Linearity:- The results obtained from a series of accuracy checks can be used to calculate the correlation coefficient and to check for linearity. The correlation coefficient is calculated by comparing the nominal and measured masses of the weights.

The acceptance criteria for the correlation coefficient shall be defined by each individual OMCL. (Proposed criterion: k = 1±0.0001).

Precision:- The precision of the balance should be verified by weighing at least 5 times a weight that is equivalent to approximately 50% of the maximum capacity of the balance. It is recommended to repeat the test with a weight that is equivalent to approximately 5% of the maximum capacity of the balance, if the balance is used at the lower range.

The acceptance criteria shall be defined by each individual OMCL.

(Proposed criterion: SD = max 5*d, where d = (actual) scale interval (e.g. d=0.1 mg)).

Eccentricity:- The eccentricity test should be carried out using a weight equivalent to at least 30% of the maximum capacity of the balance (or of the maximum weight used on the balance). The weight should be placed between halfway to ¾ of the distance from the centre of the pan to its edge and be measured at each location in the following sequence: centre, front left, back left, back right, front right, and (optional) again centre.

The acceptance criteria shall be defined by each individual OMCL.

(Proposed criterion: RSD not more than 0.05%, calculated from all weighings at different locations on the pan). 

Linearity error:- Linearity error is tested at least 3 times using four weights of defined masses whose aggregate total mass is approximately equal to half of the maximum capacity, depending on the type of balance.

First, the combined mass of all four weights is weighed and recorded. Then, two sub-sets of weights are made and their masses are recorded.

The linearity error of the balance is an absolute value calculated by the difference between the combined mass of all four weights and the sum of the masses from the two sub-sets of weights, divided by 2.

The acceptance criteria shall be defined by each individual OMCL.

(Proposed criterion: linearity error ≤ accuracy of the balances).

Drift test:- A drift test is only performed on 5, 6 or 7 decimal scale balances. It is carried out by repeated measurements (every 5 minutes for 30 minutes) of a control weight (depending on the type of balance) in the morning and afternoon.

The mean measurement, standard deviation and relative standard deviation can be calculated from the resulting data. The drift can also be extracted from the trend analysis of the verification test.

The acceptance criteria shall be defined by each individual OMCL.

(Proposed criterion: RSD = 0.05%).

Minimum weight:- The minimum weight value depends on the type of balance. It is determined from technical data and the external calibration certificate.

The acceptance criteria shall be defined by each individual OMCL.

(Proposed criterion: tolerance 0.5%, k=3 for class I balances and 1%, k=3 for class II and III balances).

Measurement uncertainty:- The measurement uncertainty can be calculated according to GUM or other relevant documents or can be determined from the external calibration certificate.

A simplified procedure is proposed as follows:

A weight of approximately 50% of the maximum capacity of the balance (depending on the type of balance) is weighed at least 10 times and the mean measurement, standard deviation and relative standard deviation are calculated. The acceptance criteria shall be defined by each individual OMCL.

Proposed criterion: measurement uncertainty is satisfactory if three times the standard deviation of not less than ten replicate weight measurements, divided by the amount weighed (approximately 50 % of the maximum capacity of the balance), does not exceed 0.001.