A calibration certificate is not calibration acceptance

What 7.1.5 asks for

7.1.5.1 asks for monitoring and measuring resources that give valid and reliable results, are suitable for the measurement and are maintained, with documented information showing fitness for purpose.

Where measurement traceability is required, 7.1.5.2 asks that equipment be calibrated or verified at specified intervals or before use against standards traceable to international or national standards, be identified so its status is known, and be protected from adjustments, damage or deterioration that would invalidate it.

The aerospace additions are where most findings occur:

• A recall process for equipment requiring calibration or verification.
• A register of monitoring and measuring equipment that includes equipment type, unique identification, location, and the calibration or verification method, frequency and acceptance criteria.
• The scope includes test hardware, test software, automated test equipment and plotters used to produce inspection data, as well as personally owned and customer-supplied equipment used to provide evidence of conformity.
• Calibration or verification carried out under suitable environmental conditions.
• When equipment is found unfit for its intended purpose, a determination of whether previous measurement results were adversely affected, and appropriate action.

The calibration evidence chain

1. Register entry
2. Recall schedule
3. External or internal calibration
4. Report review and acceptance decision
5. Status identification
6. Use
7. Impact assessment if the equipment fails

The external report is only step 3. The organization still has to decide whether the reported error is within its own acceptance criteria. If it is not, the next question is what happens to everything measured with that equipment. Laboratory competence is commonly confirmed through ISO/IEC 17025 accreditation and its scope, which many aerospace customers and Nadcap criteria also expect.

A worked example: one worn plug gauge

At fictional ABC Aerospace, an Ø8.00 plug gauge returned from calibration with 0.003 mm wear on the go end. The acceptance criterion was ≤0.002 mm.

Not notifying the customer can be the right decision. It is defensible only when the reasoning is recorded, supported by data and made by an authorized person.

A simple impact assessment flow

1. Confirm the size and direction of the error and the date of the last acceptable calibration.
2. List the part numbers, lots and characteristics judged with that equipment since then.
3. Compare the error with the tolerance. If it could accept nonconforming product, mark the lots as affected.
4. Re-measure affected product still in-house; evaluate whether shipped product requires customer notification.
5. Record the assessment, the result and who decided.

Questions an auditor may ask

• Does the register show the calibration method and acceptance criteria for this equipment? Method is the field most often missing.
• Who accepted this external calibration report, and against which criteria?
• Are operators' personal calipers on the register?
• After the last calibration failure, which product did you trace, and where is the record?
• Is your CMM measurement program treated as measuring equipment? How is it validated?

Certificate ≠ acceptance

A calibration certificate tells you how large the error is. Only the organization can decide whether that error is acceptable for its product. Leaving that judgement to the laboratory means leaving product acceptance to someone else.

Continue reading

• First-piece inspection: evidence before production release
• What makes objective evidence effective in an AS9100D audit?
• Why quarantined product can still escape