Torque Wrench Out of Calibration Mid-Job
A B1.1 engineer is torquing the cylinder hold-down nuts on a horizontally opposed engine. The specification is 45 N·m ± 5%. The calibrated torque wrench was last verified 11 months ago and carries a 12-month calibration interval. After completing 8 of 16 nuts, the engineer notices the calibration sticker is partly peeled and cannot read the due date clearly, though the paperwork in the tool crib confirms it is still within interval by 3 weeks. The wrench has a marked accuracy of ±4% of indicated value. The engineer had set it to 45 N·m and used a smooth continuous pull, but on two of the completed nuts the wrench clicked early and was re-pulled. The supervisor is under schedule pressure and suggests finishing the remaining 8 nuts and signing off.
1. Evaluate whether the completed 8 nuts can be accepted and whether the engineer may continue with the same wrench.
Read the worked answer
The wrench is within its 12-month interval, so it is not out of calibration; the unreadable sticker is an administrative defect resolved by the tool-crib record. However, the wrench accuracy of ±4% combined with the ±5% tolerance on 45 N·m means the true applied torque could lie outside the acceptance band at the extremes. The two early clicks that were re-pulled are the real concern: a click-type wrench that releases early may have been at the bottom of its scale or the nut was running down, and re-pulling a fastener already at torque can over-torque it. The engineer should not simply sign off. The correct action is to verify the wrench against a known reference or substitute a verified wrench, then re-check the two suspect nuts by backing off and re-torquing to the specified value, and record the deviation. Continuing with the same wrench for the remaining 8 nuts is acceptable only after that verification. Schedule pressure is not a technical justification.
2. State the documentation and airworthiness decision the engineer should make before release.
Read the worked answer
The engineer should raise a maintenance record entry noting the sticker defect, the tool-crib verification, and the re-torque of the two suspect fasteners. If the wrench cannot be verified on site, it must be quarantined and a replacement used, with the affected nuts reworked. The release decision rests on whether every nut is demonstrably within 45 N·m ± 5% with a traceable, in-calibration tool. If that cannot be shown, the task is not complete and the aircraft cannot be returned to service for that task. The engineer should refuse to sign a certificate of release until the evidence exists, and should escalate the tool-control weakness rather than accept the supervisor's instruction.