Design history file: assembled during the project, not after it
A DHF is a compilation, not a document. Teams that treat it as something to write at the end discover the records they needed were never created.
The certificate that only says “passed” is the one that will not help you. As-found readings and stated uncertainty are what make it evidence.
A calibration certificate records the comparison of an instrument against a reference standard: what was measured, what the instrument read, how far it deviated, and with what uncertainty. It is the evidence that the numbers your business produces can be trusted, and it is the document an auditor asks for when a measurement is questioned.
Most certificates are filed without being read, which is where the value is lost. A certificate stating only that the instrument passed cannot tell you whether it was drifting towards the limit, cannot support a decision about the interval, and cannot help when you need to know whether product measured last month was affected.
A certificate issued under an accreditation to the international laboratory competence standard carries an accreditation body mark and means the laboratory’s competence and uncertainty claims have been assessed. A certificate merely stating that the standards used were traceable to a national institute is a weaker assertion made by the supplier about itself. Both have their place; the difference matters when a customer or regulator specifies one, and it is worth knowing which you are buying rather than discovering it during an audit.
The as-found reading is what makes an out-of-tolerance event manageable. When an instrument comes back out of tolerance, the question is what it measured since the previous calibration and whether that product is affected — and answering it requires knowing how far out it was, not merely that it failed.
Calibration intervals are usually inherited from a supplier’s default or from habit. The defensible approach is to set an initial interval and then adjust it from the certificates: an instrument that is consistently well within tolerance can justify a longer interval, one that drifts needs a shorter one, and the argument for either is the history of as-found readings. That analysis is impossible if certificates are filed without extracting the data.
Ettex Records holds the certificates per instrument with the next-due date visible, Ettex Sheets tracks the as-found readings over time so drift and interval decisions are evidence-based, and the study that tells you whether the instrument is adequate for the measurement is covered in gage r and r. Where instruments are also company assets, the register is covered in it asset management.
Plainly: this is records and spreadsheets, not calibration management software, and none of it is metrology advice. Which standard applies, what uncertainty is acceptable and how a decision rule is applied are questions for your quality function and your calibration provider.
Instrument identification, as-found and as-left readings, measurement uncertainty, the reference standards and their traceability, conditions, date and the decision rule used.
They show whether the instrument was in tolerance before adjustment, which determines whether measurements taken since the last calibration are valid.
No. Accreditation to the international laboratory competence standard means the laboratory’s competence and uncertainty claims were assessed; a traceability statement is a weaker self-assertion.
From the history of as-found readings — lengthen where an instrument stays well within tolerance, shorten where it drifts.
A DHF is a compilation, not a document. Teams that treat it as something to write at the end discover the records they needed were never created.
A TMF assembled at the end is a TMF assembled from memory. Inspectors ask when documents were filed, not only whether they exist.
Most CMDB projects fail the same way — an ambitious model, a good first load, and eighteen months later a database nobody trusts enough to act on.