Appendix D Optimizing the CANBERRA iSERIES for Air Sample Counting
For the most accurate results, users should obtain standard sources that are as much
like the sample collection media as possible. If efficiency standard sources that are
unlike the samples in physical makeup or contains radioactive isotopes unlike those
likely to be found the sample, a systematic bias will be introduced to the results. This
bias will be manifest as activity results being reported that are directly related via a
ratio to the true value of the sample.
The active area of the standard is one of the input parameters available under the menu
option . Each sample counting procedure also contains similar parameter called “Unk
Sample Diameter”. When both the calibration and unknown sample diameters are the
same, the algorithm does not modify the calibration efficiency as applied to the sample
activity. However, when the Calibration Sample Diameter ? Unknown Sample
Diameter the iSERIES will modify the efficiency factor used to calculate the unknown
sample activity.
An incorrect efficiency calibration will not affect the performance of the radon/thoron
stripping algorithm; rather, an incorrect efficiency calibration will cause the calculated
activity to be off by a fixed factor. When comparing compensated results to known
values of radioactivity, it is important to understand if the values are correct,
consistently wrong, or inconsistently wrong.
Consistently Wrong — The values are not correct, but they are always off by a
consistent factor. This would indicate that the algorithm is working correctly, but
the efficiency used to correct to the activity is wrong.
Inconsistently Wrong — The values are not correct and there is no consistency
between the results and the known values. This is most likely due to a problem
with the setup leading to the algorithm working incorrectly.
Sample Distance
Like the standard source/detector distance when performing the gain calibration, the
spacing between the unknown sample and the detector is critical. If either
sample/detector spacing is not consistent between the gain calibration and the sample
counting (at 4 mm) then the algorithm will not operate in a consistent manner.
The patented
i
SERIES algorithm uses the peaks and tailings as part of the overall
subtraction routine. If the peaks do not fall into the expected region of the spectrum
(i.e. because of incorrect spacing), the algorithm will not operate correctly.
The sample/detector spacing is the most important single parameter affecting the
performance of the algorithm.
98
iMatic Automatic Counting System User's Manual - 9239221K
