Safety managers and equipment staff who run breathalyzers try to judge an instrument's condition by comparing each new calibration certificate with earlier results. Calibration is the procedure of comparing reference values with the equipment's displayed values to check performance, and the certificate contains those comparison conditions and results. But simply listing the numbers on certificates in date order makes it hard to judge whether the equipment's performance has improved or deteriorated.
For instance, even for the same instrument, this time it may have been checked over a different range of reference values than before, the display unit may have changed, or the result may come from a different stage, before or after repair or adjustment. A sound way to record repeat calibration results so they can be compared is not to turn the numbers into a single trend line, but to draw a clear boundary between results that can be compared with each other and those that cannot.
Confirm the identity of the instrument before the numbers
The starting point for comparison is the model name and serial number. Even breathalyzers of the same type used in the same place must be recorded as separate instruments if their serial numbers differ. If an instrument was replaced or handed over from another department, you also need to confirm first whether the earlier certificates are part of the current instrument's history.
Even if the identification matches, if the configuration changed, it is hard to read the results as a continuous series under the same conditions. If work such as replacing the sensor or related parts, repair, or adjustment took place, record the date and nature of the work next to the result. This marker is not a stigma indicating a problem; it is a boundary line needed to interpret differences in the numbers before and after.
Record comparison criteria in a single table
Rather than copying every item from the certificate, it is better to organize the conditions needed to compare rounds in the same order each time. In particular, a calibration point indicates the reference value at which the instrument was checked, so the results of two rounds with different calibration points must not simply be expressed as an increase or decrease. Write down units and test conditions based on the original certificate, and do not fill them in from memory or habit.
| Comparison item | What to record | How to handle differences |
|---|---|---|
| Instrument | Model name, serial number, control number | Keep a different instrument as a separate history |
| Comparison conditions | Calibration points, display unit, test conditions stated in the certificate | Do not compare directly if they are not the same |
| Stage of the result | Whether it is before or after repair or adjustment, whether parts were replaced | Mark the date and nature of the change next to the result |
| Source document | Certificate number, issue date, where the original is kept | If the original is missing, record the meaning of the figures as unconfirmed |
For example, if an older certificate has results for only one reference value and a recent certificate has results for several, you can make a limited comparison using only the reference values the two share. If you average in the items newly added in the recent certificate, the measurement range itself changes and that can look like a change in performance. Do not delete rounds or items excluded from comparison; recording the reason for exclusion alongside them increases the reliability of the records.
Do not mix before-and-after adjustment results in one row
Whether adjustment or repair took place during calibration has a large effect on how results are interpreted. A before-adjustment result can show the instrument's condition when it was received, while an after-adjustment result can indicate the condition confirmed after the work. If you put both results in the same column and compare them with the previous round, you will confuse the instrument's change over time with the effect of the work.
Consider a hypothetical case. Suppose the management log lists figures from 2024 and 2025 certificates, and you later discover a note indicating that the 2025 certificate shows after-adjustment results. The facts to observe are that the two figures differ and that work was done before the 2025 result. Rather than concluding that the instrument's performance improved, the appropriate action is to separate the 2025 result as an "after-adjustment confirmed value" and to check the before-adjustment result or the nature of the work with the body that issued the certificate. The completion criterion is a state where you have confirmed whether the two rounds share the same calibration points, units, and result stage, or have clearly recorded any differences that cannot be confirmed as reasons for excluding them from comparison.
The absence of certificates from both before and after a repair does not let you assume the instrument was in normal condition during that period. You need to record separately the scope shown by the confirmed documents and your internal review opinion. If you record the original result using the certificate's own wording and enter judgments such as "comparable" or "not compared directly due to different conditions" in a separate column, the next person in charge can also tell the evidence from the interpretation.
Do not fill blanks with past values
Sometimes an old equipment register contains only numbers, with no original certificate. In that case, if you cannot tell what unit a number is in, which reference value it came from, or whether it is a before- or after-adjustment result, it is better not to fill the blanks with information from an earlier round. Supplementing them with a certificate for a similar model or the conditions of another instrument changes the source of that number.
If you can find the original document, link it to the certificate number and issue date, and if the description of conditions is insufficient, you can ask the issuing body what the result means. When an answer changes your internal interpretation, record the date and reason for the change rather than erasing the earlier opinion. The purpose of history management is not a neat continuous graph, but making it possible to recheck the basis for judgments later.
Keep model-specific materials and review dates together
If the instrument's model or the applicable materials change, the explanations referred to in the comparison can change too. DATECH's Downloads page offers the Korean manuals for The New AF-50 and the AF-50AD as separate materials. So if you record the title and date checked of the manual or manufacturer material used when reviewing certificates, it is easier to explain the background to your judgments even if the materials change later or someone else does the review.
Repeat calibration history can be material for reviewing future management intervals, but the fact that results were similar a few times is not by itself a reason to automatically lengthen the interval. ILAC-G24:2022/OIML D 10:2022 also points toward keeping calibration history as material for judging future intervals. Only by building up records that distinguish comparable rounds, rounds put on hold because conditions differed, and rounds divided by a repair or adjustment boundary can the next review look at the numbers and their context together.
