FISCHER FISCHERSCOPE X-RAY 5000 Series, Руководство оператора

Страница: 105 / 188

Wouldn’t it be better to re-calibrate? Measurement device monitoring for the Fischerscope X-RAY
FISCHERSCOPE
®
X-RAY 105
ment device monitoring is not recommended directly
after power-up of the instrument, even though regular
measurements can be made.
14.9 Wouldn’t it be better to re-calibrate?
To correct a measurement application that does no longer ”fit”
according to the control chart, a reference measurement and a
normalization or the verification and possibly but very rarely, a re-
measurement of the spectra library are generally sufficient. The
calibration of the measurement application is not changed. The
verification of the measurement application that is to be performed
after the correction will indicate if this is sufficient. Our experiences
up to now has shown, repeating a calibration is almost never
necessary.
Still, various customers have performed re-calibrations of their
measurement applications without sufficient measurement device
monitoring as recommended by us due to erroneously postulated
(although well-meaning) procedural instructions (test and process
instructions) in their quality management manual. They have only
done this because supposedly a re-calibration must be performed
after a certain period, e.g., after one year of instrument operation.
We can only advise against such a procedure. With this method, the
user knows nothing about the current situation of the instrument,
neither before nor after this measure, and can also not demonstrate
proper measurement device monitoring. Users that have followed
our recommendations to this point know what we mean.
In practical applications, full re-calibrations have at most proven
necessary after replacing the tube or the detector. On the other
hand, each new calibration measurement will (just as any regular
measurement) lead to unavoidable random deviations with regard
to the old measurement results. With each calibration, a result
afflicted with random deviations is ”frozen” in place such that no
new calibration is equal to the previous one. Yet, the calibration is
the foundation for all subsequent measurement results. The
measurement results change (albeit slightly) after re-calibration.
Re-calibrations can also necessitate a pre-run with re-calculation of
the control limits of the quality control chart that is used to check
the measurement application because the old control limits no
longer fit to the new calibration. Thus, instead of a new calibration
we recommend as a corrective measure only a reference
measurement and normalization, or a spectra check, respectively,
as long as the verification measurement following the corrective
measure indicates that the readings are still okay. It is better to