Calibration Troubleshooting
A passed calibration certificate does not prove that a part will pass inspection. When the two results seem to conflict, review the instrument, inspection method, specification, measurement uncertainty, environment, part, and production process before drawing a conclusion.

Your Calibration Certificate Passed. Why Is the Part Still Failing Inspection?
A valid calibration certificate does not guarantee that every part measured with the instrument will pass inspection. It only provides evidence about the instrument’s performance at the time of calibration, within the ranges and conditions shown on the certificate.
Inspection answers a different question: does this specific part meet the applicable requirement?
There may be no conflict at all. The instrument could be working correctly while the part is genuinely outside tolerance. The result could also be affected by the inspection method, instrument setup, environmental conditions, measurement uncertainty, an incorrect specification, or the condition of the part.
Instead of treating a passed certificate as proof that the inspection result must be wrong, examine the complete measurement process.
Calibration and Inspection Answer Different Questions
Calibration establishes the relationship between an instrument’s indications and recognized measurement standards under stated conditions. Inspection uses a measurement result to decide whether a part satisfies a documented requirement.
An acceptable calibration result does not prove that:
- The instrument is suitable for every measurement task.
- The inspection method was followed correctly.
- Different operators will obtain equivalent results.
- The instrument has remained unchanged since calibration.
- The correct specification or decision rule was applied.
- The part meets its requirements.
Do not rely on the word “pass” alone. Review the certificate’s reported values, calibration points, covered range, stated conditions, uncertainty, and any limitations or remarks. Then decide whether that evidence applies to the measurement in question.
Check the Instrument
First, confirm that the instrument used for inspection is the instrument identified on the calibration record. Match its serial number, asset number, or other unique identifier.
Review its calibration status, relevant results, measurement range, resolution, and physical condition. Check any intermediate verification, reference check, or performance-check records required by your quality system.
Make sure the inspection was conducted within the range and conditions covered by the calibration. An instrument may perform acceptably at one range or calibration point but lack the resolution, capability, or supporting evidence needed for another tolerance or application.
Verify the complete configuration, including units, operating mode, software settings, probes, fixtures, accessories, correction factors, and compensation features. A valid certificate cannot prevent an incorrect setup from producing a misleading result.
Also review the instrument’s history since calibration. A drop, overload, adjustment, repair, software change, improper transport, contamination, or exposure to unsuitable conditions may mean the certificate no longer represents its current performance. If there is reason to doubt the instrument, remove it from service and follow the organization’s verification, impact-assessment, or recalibration procedure.
Review the Inspection Method
Next, examine exactly how the part was measured.
Even with the same calibrated instrument, small differences in technique can change the result. Review the measurement location, part orientation, alignment, datum structure, reference surfaces, contact force, fixture, measurement sequence, and operator instructions.
Environmental conditions may matter as well. Temperature, humidity, vibration, cleanliness, handling, and stabilization time can influence a measurement. Their significance depends on the part, instrument, method, material, and tolerance.
Two readings may not be comparable if operators measure different locations, position the part differently, use different fixtures, or establish different reference points. Calibration controls only part of the measurement process; it does not remove variation introduced by the application.
When practical, repeat the inspection under controlled conditions using the approved method. Record enough detail for another qualified person to reproduce the measurement. The purpose is to test a defined cause, not to repeat measurements until one happens to pass.
Check the Specification
Confirm the exact requirement used to evaluate the part.
Use the current approved revision of the drawing, specification, customer requirement, inspection plan, or procedure. Record the actual tolerance and acceptance criteria rather than relying on memory, a nominal value, or an expected result.
Make sure the characteristic measured is the characteristic defined by the requirement. Similar dimensions or features may use different datums, units, measurement definitions, conditions, or acceptance rules.
Common specification problems include:
- Incorrect units or conversion errors.
- An outdated drawing or specification revision.
- Transcription, programming, or data-entry errors.
- Unapproved tolerance changes.
- Different requirements for similar part variants.
- Special notes that override a general tolerance.
- Requirements that apply only under stated conditions.
The nominal value alone does not determine acceptance. The approved requirement and applicable decision rule control the conformity decision.
Consider Measurement Uncertainty
Measurement uncertainty deserves close attention when a result is near a specification limit.
Uncertainty describes the range of values that can reasonably be attributed to the quantity being measured, based on the available information. It is not simply a warning that the instrument could be wrong. It reflects relevant sources of variation associated with the full measurement process.
Depending on the task, those sources may include calibration uncertainty, instrument resolution, repeatability, reproducibility, environmental effects, reference standards, alignment, fixtures, part condition, operator technique, and the inspection method.
Before declaring a near-limit result conforming or nonconforming, identify the required decision rule. It may be set by a specification, standard, customer agreement, contract, or approved internal procedure.
Do not apply a generic guard band or acceptance formula to every inspection. The correct treatment of uncertainty depends on the requirement governing the decision.
Document the decision rule and the role of uncertainty clearly enough that another qualified person can understand and reproduce the conformity decision.
Investigate the Part and Production Process
A failed inspection may reflect a real change in the part or process.
Compare the current result with earlier inspection records and relevant production data. Depending on the process, useful evidence may include machine settings, tooling changes, tool-wear records, maintenance activity, material records, environmental data, operator changes, and documented process adjustments.
Look for repeatable patterns. Does the shift appear across several parts? Did it begin after a recorded process event? Does it occur only with one instrument, fixture, operator, machine, batch, or inspection location?
Timing can point you toward useful evidence, but it does not establish cause. Do not blame a process change without supporting data. The same applies to the measuring equipment: one failed part does not prove that the instrument is defective.
Compare Results Carefully
If the result conflicts with an earlier measurement, determine whether the two results are truly comparable.
Confirm that both measurements used:
- The same measurand.
- The same units.
- The same measurement location.
- Equivalent methods and fixtures.
- The same datum or reference structure.
- Comparable environmental conditions.
- Equipment suitable for the required tolerance.
- The same specification revision.
- Compatible decision rules.
A difference between readings does not automatically mean one result is wrong. The methods may define or access the characteristic differently, or the difference may be reasonable once repeatability and measurement uncertainty are considered.
Conduct controlled repeat measurements when the evidence supports doing so. Avoid changing instruments, operators, locations, or methods repeatedly until a passing result appears. That approach can hide the cause of the discrepancy and undermine the inspection record.
Document the Investigation
Keep the evidence and reasoning together so the investigation can be reviewed later.
Record the instrument identification, calibration status, relevant certificate results, measured value, inspection method, environmental conditions, applicable specification, uncertainty where relevant, decision rule, supporting production records, findings, and final disposition.
The record should allow another qualified person to determine:
- What was measured.
- Which requirement applied.
- How the measurement was performed.
- What evidence was reviewed.
- How the conformity decision was reached.
If the investigation results in corrective action, record the action, assign an owner, and verify that it worked. Actions might include revising an inspection instruction, repairing or checking a fixture, clarifying a drawing requirement, retraining an operator, improving environmental controls, or addressing a production problem.
Inspection Investigation Checklist
Use this checklist when an instrument passes calibration but a part fails inspection:
- Confirm the instrument’s identity and current calibration status.
- Review the calibration results that apply to the measurement.
- Check the instrument’s range, resolution, condition, and configuration.
- Review required intermediate checks or performance verifications.
- Confirm that the approved inspection method was followed.
- Check the measurement location, positioning, and alignment.
- Review reference surfaces, fixtures, contact conditions, and sequence.
- Confirm that required environmental conditions were maintained.
- Verify the current specification revision and stated tolerance.
- Confirm that the correct characteristic was measured.
- Assess measurement uncertainty when the result is near a limit.
- Identify and apply the required decision rule.
- Compare relevant part, inspection, and production records.
- Repeat the measurement under controlled conditions when justified.
- Record the evidence, findings, disposition, owner, and corrective action.
- Verify and document the result after any change is made.
Continue the Investigation
If the instrument has a valid calibration certificate but the result still appears unreliable, read Your Instrument Is Calibrated. Why Can the Measurement Still Be Wrong?
If two calibrated instruments produce different readings, see Two Calibrated Instruments Give Different Readings. Which One Should You Trust?
If recalibration did not resolve the problem, read The Instrument Was Recalibrated, but the Measurement Problem Remains.
If the result is only slightly outside tolerance, see A Measurement Is Just Outside Tolerance. Should You Reject the Part?
A passed calibration certificate is useful evidence, but it is only one part of the conformity decision. Review the instrument, method, environment, specification, uncertainty, part, and production process before deciding why the inspection failed.