National Physical Laboratory

Assessment of uncertainty in a measurement

Question BinocularsEvery measurement has an uncertainty associated with it.

A calibrated instrument has an uncertainty attached to its calibration factor; this can be obtained from the calibration certificate.

If  the instrument is used to make a measurement in identical conditions to those in which it was calibrated then only the uncertainty in the calibration factor will need to be considered. However usually measurements will be made in conditions vastly different to those in the test facility; this will increase the uncertainty in the reading.

The uncertainty for a radiation survey or contamination survey result needs to include:

  • uncertainty in the instrument calibration factor

AND

  • uncertainty arising from the monitoring process

Generally, the uncertainty in the instrument calibration factor will be assessed by a specialist test house using well-defined sources and measurement facilities and the uncertainty on the calibration factors may be as small as a few percent.

However, the monitoring process is very much less well-defined and the associated uncertainties are usually at least one order of magnitude greater than the calibration factor uncertainty and may often be much larger.

The following sources of uncertainty may need to be considered:

Dose rate monitors

Most likely deviations from calibration conditions may be:

  • The energy of the incident radiation is frequently unknown, due to scatter and absorption. As a consequence of this, the uncertainty of the response of the instrument can be up to ±30 %.
  • The directional distribution of the radiation field.
  • Uniformity of the radiation field.
  • Pressure and temperature effects for vented ion chambers (unless accounted and corrected for).

Additional uncertainties associated with the measurement may be:

  • statistical variation of the observed reading
  • dead time
  • area over which the measurement is averaged
  • variation in radiation background level. 

Contamination monitors

Most likely deviations from calibration conditions may be:

  • Energy of the incident radiation. This is unlikely to be mono-energetic or of the same energy as the calibration energy.
  • Condition of the contaminated surface. The type of surface, the amounts of dust, grease and grime covering or mixed in with the contaminant and the shape of the surface.
  • Distance of the detector or probe from the contaminated surface.
  • Speed of monitoring, i.e. how fast the detector was moved over the surface being monitored.
  • Distribution of contamination, i.e. whether the contaminant is distributed as random hot spots or is uniformly spread across the surface.
  • Direction of the radiations emitted.

Wipe test measurement

Uncertainties will be due to:

  • type of wipe used
  • pressure applied by the person when making the wipe
  • area wiped
  • contamination distribution
  • porosity, chemical composition, texture and cleanliness of the surface.

The amount that a wipe removes, or the ‘Pick-up’ factor, varies enormously and is almost impossible to assess accurately. The uncertainties in the Pick-up factor are an order of magnitude larger than other uncertainties in the measurement.

 

In any measurement scenario, a record should be kept of which factors have been considered that contribute to the uncertainty along with any assumptions made.

A significant quantity of detailed guidance on uncertainties generally and on ionising radiation measurements in particular is available 

Arrow Left

1 of 2Arrow Right