ISO/IEC 17025 clause 6.4.9: control faulty or questionable equipment
Clause 6.4.9 requires equipment to be taken out of service when overload, mishandling, questionable results, defects or failure to meet specified requirements make it unreliable. Prevent its unintended use, examine the effect of the problem and initiate the nonconforming-work process. The duty extends beyond arranging a repair.
What clause 6.4.9 requires in practice
Clause 6.4.9 describes five situations in which equipment must stop being used: overloading, mishandling, questionable results, evidence of a defect, and evidence that it is outside specified requirements. In any of these, the item is withdrawn from use and either kept apart or plainly marked as not available, and it stays that way until checks confirm it performs properly.
The laboratory must also examine what the defect or deviation did to earlier work and start its procedure for nonconforming work, which is set out in clause 7.10. That cross-reference shows that the clause is as much about results as about the instrument.
The clause does not dictate how isolation is achieved; a tag suits a portable item while access restriction may suit an installed system. It also does not say all previous results are invalid, only that their possible exposure must be examined.
The outputs are visible containment, an evaluation of affected work, a verified return to service and a record of what was decided.
Control use immediately and visibly
Isolate the equipment or clearly identify it as out of service until correct performance has been verified. The method must work in the laboratory's circumstances. A tag may be appropriate for one item; access restrictions or another effective measure may be needed for equipment that remains physically installed.
Tell the people who could use the item or its data. Consider connected systems and automated workflows, not only the person who found the problem. Reassigning work to another instrument also requires suitable equipment and competent personnel; urgency does not remove those controls.
Evaluate the effect on results
Determine what failed, when the problem may have begun and which activities or results could be affected. Use available calibration history, intermediate checks, maintenance records and technical observations. Avoid assuming that every previous result is invalid, but do not assume that only the last result matters either.
Clause 7.10 provides the process for significance evaluation, decisions about acceptability, customer notification or recall where necessary, and authority to resume work. The investigation should support those decisions. A technically justified boundary for affected work is more useful than a blanket statement that “no impact was found” without evidence.
Verify before returning to service
After repair or adjustment, establish correct performance for the intended use. Consider the relevant verification, calibration, correction information and software settings. A service technician's statement that the instrument turns on is not necessarily evidence that the laboratory's measurement requirements are met.
Update the equipment records and status. Where the problem may recur or indicates a wider system weakness, use the appropriate corrective-action arrangements. Restoration of one instrument does not automatically address poor handling, inadequate checks or an unsuitable calibration programme.
Why control of doubtful equipment matters to the business
A faulty instrument is dangerous mainly because of what it did before anyone noticed. Results already reported may need to be revisited, customers may need to be told and products may already have been released on the strength of those numbers.
Fast, visible containment limits the number of results at risk, and a careful evaluation limits both overreaction and underreaction. Laboratories that handle this well can speak openly with customers about what happened, which protects trust. Those that rely on an informal sense that results are probably fine are exposed when the next assessor or customer asks to see the reasoning.
How to implement clause 6.4.9
Prepare before the first failure, because the hours after discovery are when decisions are rushed.
- Agree the triggers that lead to withdrawal and tell staff that anyone may raise them.
- Define how an item is isolated or marked in your premises, including installed equipment and connected systems.
- Tell everyone who could use the item or its data, and check that replacement equipment and people are suitable.
- Establish the earliest point at which the problem may have begun, using calibration history, intermediate checks and maintenance records.
- List the jobs that used the item since that point and pass them into the nonconforming-work process for a significance decision.
- Verify performance for the intended use after repair, covering calibration, correction information and settings where relevant.
- Update records and status, and consider whether similar items or handling practices need review.
- Practise the response with a simulated failure from time to time, so that staff know the steps and the decision makers are already identified.
Two illustrative equipment incidents
A reference instrument is dropped during transport to a customer site. The team removes it from use and informs the responsible technical person. It evaluates whether any measurements were taken after the event and what verification is needed. The absence of visible damage is not treated as proof of correct performance.
A testing laboratory detects a balance outside its acceptance criteria. It reviews the relevant checks, calibration information and jobs using that balance to determine possible impact. If customer results may require action, it follows the nonconforming-work process rather than closing the incident after adjustment.
Evidence and mistakes to avoid
Retain relevant equipment history and nonconforming-work records showing the event, containment, technical evaluation, decisions and release. Clause 6.4.9 does not prescribe a single incident-form design. The evidence needs to explain what happened and why the resulting decisions were justified.
Common weaknesses include leaving the item available while waiting for repair, failing to assess previous results, using an undocumented “probably fine” judgement, or returning equipment without verification. A failure can also reveal a need to review similar equipment or handling practices; consider that possibility through corrective action where appropriate.
Frequently asked questions
Must every equipment defect lead to withdrawal of all previous reports?
No. Evaluate the significance and affected work technically, then take the actions required by the nonconforming-work process. Neither automatic recall nor automatic dismissal is appropriate.
Can we keep using the instrument while a service visit is arranged?
Equipment meeting the clause 6.4.9 triggers must be taken out of service and prevented from unintended use until correct performance has been verified.
Is repair evidence enough to resume work?
Not necessarily. Verify correct performance for the intended laboratory use and follow the defined authority for release and work resumption, including relevant calibration requirements.
Turn this requirement into reliable laboratory practice
Sterling can help examine your equipment-failure response from the first warning to customer-result decisions. Bring an anonymised example to a laboratory consultation to identify weaknesses in containment, technical review and follow-up.
Build the team's understanding through Nonconformity Management. Discuss the relevant scope, employee learning and agreed access to Effivity ISO & GRC automation software with Sterling; laboratory technical decisions remain with competent, authorised personnel.
ISO/IEC 17025 requirements hub · Equipment controls · Nonconforming work · Corrective action