1.2.2. Types of models¶
Regarding the relation between the output quantity and the net count rate, or the activity in the counting source in the case of linear unfolding, three types of models of measurement are considered in UncertRadio:
positive linear model, including detection limit calculation
negative linear model, including detection limit calculation
only GUM, without detection limit calculation
The relation described in the section above, represents the most often encountered case of a positive linear model. In this case, the net count rate \(R_{n}\) increases with an increasing gross count rate. An activity is detected, if the gross count rate is significantly larger than the background count rate.
Negative linear Model
With the measurement model explained above, the net count rate Rn increases with increasing gross count rate; it is called positive linear. The activity is considered as detected if the gross count rate significantly exceeds the background count rate.
A negative linear model (since version 2.1.10 available) is characterized by a gross effect which must fall significantly below the “background” for the detection of the effect. In this case the difference in the expression \(R_{n}\) for the net effect is reversed:
\(R_{b}\) and \(R_{0}\) herein are not necessarily count rates. The project Rn-222-Emanation_EN.txp is an example for it.
The case abbreviated as “only GUM” has only been introduced for the situation that only an uncertainty according to GUM is of interest for a measurement, but a detection limit shall not be calculated, as it occurs for instance in the case of weighing a sample.
