Expired Balance Calibration and an Unverified Check Weight
An environmental laboratory determines cadmium in soil digests by graphite furnace atomic absorption. Sample weights of about 0.2 g are measured on a balance with 0.01 mg readability. The balance was calibrated 13 months ago by an ISO/IEC 17025-accredited calibration laboratory; the interval in the equipment register is 12 months. The certificate states that the stated uncertainty is valid for 20 degrees C plus or minus 2 degrees, and the weighing room log shows 24 to 27 degrees from June to September. Every morning a technician checks the balance with a 100 mg E2 check weight; readings have stayed within plus or minus 0.05 mg, and the check is written on a clipboard. The check weight's own certificate is four years old and no recalibration is scheduled. The uncertainty budget for cadmium lists only balance resolution (0.01 mg) and weighing repeatability. It does not mention the balance calibration uncertainty, the check weight, or temperature.
1. List the requirements that are not being met and explain how each failure interacts with the others.
Read the worked answer
Four requirements are unmet, and they compound each other.
Calibration interval (clause 6.4). The laboratory's own register sets 12 months and the last calibration was 13 months ago. Since month 12 the balance has been used outside the programme the laboratory itself defined - a departure from its own arrangements, which is a nonconformity regardless of whether the balance happens to be accurate.
Validity of the calibration result. The certificate's uncertainty is stated for 20 plus or minus 2 degrees. The room is logged at 24 to 27 degrees in summer. The calibration therefore does not describe the balance under the conditions of use, and the environmental conditions requirement (clause 6.3) is implicated because a known influencing quantity was neither controlled nor corrected for.
Traceability (clause 6.5). The check weight's certificate is four years old with no recalibration scheduled. The daily check therefore rests on an unverified value, so the chain of calibration is broken at the check as well as at the balance.
Uncertainty budget (clause 7.6). Resolution and repeatability only. Balance calibration uncertainty, the check weight's contribution and the temperature effect are all missing, so reported uncertainty is understated. That matters because cadmium results are compared against limits, where an understated uncertainty changes the conformity decision.
2. The laboratory argues that the daily check has passed every day, so the balance is demonstrably performing and the overdue calibration is a paperwork issue. Evaluate that argument.
Read the worked answer
Reject it, for three separate reasons.
Function. An intermediate check detects drift between calibrations; it does not establish traceability and does not extend a calibration interval. Traceability comes from an unbroken chain of calibrations, and that chain has an expired link.
Coverage. The check is a single point at 100 mg, while the balance is used near 0.2 g. A single-point check says nothing about span or linearity at the working point, and nothing about behaviour when the room is 5 degrees above the calibration condition.
Independence of the check. The check weight's own value is four years old and its uncertainty is unknown. A pass means only that the balance has not moved relative to an unverified artefact - which is not reassurance, it is circular.
There is also a design question: plus or minus 0.05 mg at 100 mg on a 0.01 mg readability balance is a loose limit. A check that always passes may mean the limit was set to be easy to pass rather than to detect a defined change. The honest position is that results produced during the affected period carry unquantified uncertainty, which is a technical nonconformity rather than an administrative one.
3. Specify what the laboratory must do before resuming cadmium determinations on this balance, and what it owes to results already reported.
Read the worked answer
Before resuming. Take the balance out of service and label it. Recalibrate it either in situ at the actual working temperature or after bringing the room under control at 20 plus or minus 2 degrees and demonstrating stability - the choice depends on whether the laboratory can realistically control the room. Recalibrate or replace the check weight and bring it into the equipment programme with its own interval. Rewrite the uncertainty budget with all relevant components: calibration uncertainty, resolution, repeatability, check weight, and temperature sensitivity, combined with a stated coverage factor.
Retrospective impact assessment. Identify every batch weighed while the calibration was out of interval or outside its stated temperature range. Estimate the effect on sample mass, propagate it into the cadmium result, and compare with the decision threshold for each affected report. This is the significance evaluation required for nonconforming work: the question is whether any reported conclusion changes, not whether anybody noticed.
Obligations. If results are affected, notify the client in writing and recall or amend the reports. Record the impact assessment, the notification and the technical records that allow recalculation. Raise a corrective action (clause 8.7) with a real root cause - calibrating equipment without linking its validity conditions to the room's actual environment, and managing the check weight outside the calibration programme - then update the risk register and verify effectiveness at the next audit.