The performance statistic most familiar to feeder users, repeatability quantifies the degree to which a feeder’s discharge stream varies over brief time intervals, producing a snapshot of one dimension of feeder performance.
Note that the repeatability measurement says nothing at all about whether the feeder is delivering, on the average, the targeted rate (that is the purpose of the linearity measure performed on a properly calibrated feeder). Repeatability does, however, reveal a great deal about the expected extent of short-term flow-rate inconsistencies, an important contributor to the integrity of the formulation and end-product quality.
The repeatability measurement is performed by taking a series (usually at least 30) of carefully timed consecutive catch samples from the discharge stream, weighing each, and then calculating the +/- standard deviation of sample weights expressed as a percentage of the mean value of the samples taken. The measurement is typically performed at the nominal intended operating rate(s) of the feeder. See measurement procedures section for more information.
For example, owing to the random nature of repeatability errors, if sampling shows a standard deviation of +/- 0.3% it can be said that 68.3% of sample weights will fall within the +/- 0.3% error band (1 Sigma), 95.5% will occur within +/- 0.6% (2 Sigma), and 99.7% will lie within +/- 0.9% (3 Sigma). These expressions are equivalent.
Traditionally, repeatability has been expressed at two standard deviations (2 Sigma) over minute-to-minute sample periods. However, due to higher throughput rates and more stringent quality standards, many processors are now requiring sampling periods as short as several seconds. Where such short sampling periods are required, a corresponding lowering of repeatability precision is to be expected. For more information, see the following section on repeatability timescales.
A complete expression of a repeatability statistic must contain the following elements: a +/- percentage error value, the Sigma level, and the sampling criteria. For example, a repeatability performance statement might take the following form: +/- 0.5% of sample average (@ 2 Sigma) based on 30 consecutive samples of one minute, one kilogram, one belt revolution, or thirty screw revolutions, whichever is greater.