How to Receive and Validate a Checking Fixture: Dimensional Certificate, Gage R&R and Acceptance Criteria

The checking fixture has arrived at the plant and the question is whether you can release it to the line. Only when you have three pieces of evidence: a physical inspection with no damage, a dimensional certificate traceable to the CAD, and a measurement system study (Gage R&R or attribute agreement) that meets the AIAG MSA guidelines. Here is how to check each one and what to ask for in the purchase order.

In Summary

• A checking fixture is a measurement system: before releasing it to the line you have to show that it measures well, not just that it "looks right".
• Receiving has three layers: physical inspection, a traceable dimensional certificate and a measurement system capability study (Gage R&R or attribute agreement).
• The AIAG MSA guidelines are the practical reference in automotive: %GRR < 10% generally acceptable, 10–30% conditional, > 30% unacceptable, and ndc ≥ 5, without using them as the only criterion.
• Agree on these deliverables in the purchase order, not once the fixture has already arrived; in Radii tooling and MRO you can ask for them from the quote.

If you are deciding when a custom fixture makes sense versus a catalog one, we cover that in dimensional inspection fixtures: custom vs catalog. This article is the next step: the fixture exists and you have to accept it.

1. Why receiving is not a formality

A checking fixture is a monitoring and measuring resource (clause 7.1.5 of ISO 9001:2015, which IATF 16949 incorporates). ISO 9001 (7.1.5.2) requires that, when traceability is a requirement, the equipment be calibrated or verified at specified intervals or before use against standards traceable to national or international standards. IATF 16949 adds two things: statistical studies of the variation of each type of inspection, measurement and test system identified in the control plan, using the methods and criteria of the MSA reference manuals unless the customer approves others (7.1.5.1.1), and a documented process for the calibration and verification records of all gauges (7.1.5.2.1).

In practice, if you release a fixture without evidence and an audit or a complaint later shows that the fixture's own variation was comparable to the part tolerance, everything accepted with it is in doubt.

2. Physical inspection on arrival

Before measuring anything:

1. Identification: fixture number, part number and the engineering level it checks, engraved or on a plate.
2. Integrity: no dents on support surfaces, locators and pins free of burrs and marks, clamps that close without play.
3. Master or reference part: if agreed, it arrives identified and with its own measurement report.
4. Match with the design: the fixture corresponds to the approved model or drawing (same datums, same support points as the part's GD&T).
5. Accessories: go/no-go gauges, flags and indicators, with identification and calibration status.

If the locating pins show wear or damage on delivery, stop here.

3. How to read the dimensional certificate

The certificate should let you answer four questions:

A common practice in metrology is that the instrument's uncertainty or tolerance be much smaller than the tolerance being checked (the 10:1 rule we discuss in the custom vs catalog article). If the report does not state the equipment or its calibration, you cannot confirm it.

4. Gage R&R: when and how

If the fixture delivers values (indicators, sensors, caliper measurement on fixture points), the applicable study is a variable Gage R&R. A frequent design is 10 parts, 3 appraisers and 2 or 3 trials, with parts that represent the real process variation. Confirm with your customer which design they require.

AIAG MSA (4th ed.) guidelines for variable %GRR:

• %GRR < 10%: generally acceptable.
• 10% to 30%: may be acceptable for some applications depending on importance, cost and risk; agree on it with the customer.
• > 30%: unacceptable; the fixture or the method has to be improved.
• ndc ≥ 5: the system distinguishes enough categories within the process variation.

Two nuances: make clear whether %GRR was calculated against the total variation of the study or against the tolerance, because they answer different questions; and the manual itself warns that using these thresholds alone is not enough to accept a measurement system.

What to check in the result:

• If repeatability (equipment variation) dominates, it usually points to the fixture: play in the clamps, worn locators, unstable support.
• If reproducibility (variation between appraisers) dominates, it usually points to the method: missing loading instruction, clamp sequence, clamping force.

Illustration: diagram made by Radii with the AIAG MSA (4th ed.) %GRR and ndc thresholds and the ISO 9001 7.1.5.2 and IATF 16949 7.1.5.1.1 clauses cited in this article.

5. Go/no-go fixtures: attribute agreement

Many checking fixtures give an attribute result: the pin enters or it does not, the flag sits inside or outside. There, variable %GRR is not the right analysis. You use an attribute agreement analysis: a set of parts with a known reference value (including parts close to the limits, which is where appraisers fail) is evaluated several times by several appraisers, and you measure each appraiser's agreement with themselves, with each other and against the reference. A common design is about 50 parts, 3 appraisers and 3 trials, and the result is usually expressed with kappa: as a guideline, AIAG considers 0.75 or more good agreement and below 0.40 poor. The customer may set stricter criteria.

6. Release and afterwards

With inspection, certificate and study in order:

1. Register the fixture in the measuring equipment control system, with its calibration interval (the one your quality system or the customer defines).
2. Post the work instruction at the point of use.
3. Define what happens if the fixture is hit or modified: re-verification before it is used again.
4. Keep the package as part of the evidence for the PPAP and control plan.

7. What to ask for in the purchase order

To avoid arguing about deliverables at the end, specify them from the quote:

• Dimensional certificate of the fixture against CAD, with the measuring equipment identified and its calibration current.
• 3D model and fixture drawings as deliverables.
• Gage R&R or attribute agreement study (who runs it, with which parts and against which acceptance criterion).
• Work instruction.
• Master part, if applicable.

If you are quoting fixtures, tooling or spare parts, in tooling and MRO you can upload your drawings and ask for these deliverables to be part of the quote, or start directly at app.radii.com.mx.

Frequently Asked Questions

What should a checking fixture delivery package include?

At a minimum: the fixture's dimensional certificate against its CAD (reference points, locators, support surfaces), identification of the measuring equipment used and its current calibration, drawings or 3D model of the fixture, work instruction and, if agreed, the Gage R&R or attribute agreement study.

What %GRR is acceptable for a checking fixture?

The AIAG MSA reference manual (4th edition) guidelines are usually summarized like this: %GRR under 10% is generally acceptable; between 10% and 30% may be acceptable for some applications depending on importance, cost and risk; over 30% is unacceptable and the measurement system must be improved. A number of distinct categories (ndc) of 5 or more is also recommended. The manual itself warns that these thresholds should not be used alone as the acceptance criterion, and the customer may set others.

How do you validate a fixture with go/no-go gauges?

When the result is an attribute (go/no-go, flag, pin that enters or does not), variable %GRR is not the right analysis; you use an attribute agreement analysis with reference parts of known value, including parts close to the limits, and evaluate each appraiser's agreement with themselves, with the others and against the reference, typically with the kappa statistic. As a guideline, AIAG considers a kappa of 0.75 or more good agreement and below 0.40 poor.

How often should a checking fixture be recalibrated?

There is no universal interval. The plant's or the customer's quality system defines it based on use, wear and risk. ISO 9001:2015 (7.1.5.2), which IATF 16949 incorporates, requires calibrating or verifying at specified intervals or before use against traceable standards, and IATF 16949 (7.1.5.2.1) requires a documented process for the calibration and verification records of all gauges.

---

Conclusion: a fixture without evidence is not released yet

• Inspect the fixture physically before measuring anything.
• Require the dimensional certificate against CAD, with the measuring equipment identified and its calibration traceable.
• Run the Gage R&R or the attribute agreement study and judge it with the AIAG MSA guidelines, without using them as the only criterion.
• Agree on all of it in the purchase order, not on arrival.

At Radii, when you quote a fixture, tooling or a spare part, you can ask for the dimensional report to be part of the delivery. Upload your drawing or CAD at app.radii.com.mx.

Sources

• AIAG, Measurement Systems Analysis (MSA) Reference Manual, 4th edition (2010). %GRR, ndc and kappa guidelines as summarized publicly by SPC for Excel, Minitab and IJQR, "On the relevance of ndc".
• IATF 16949:2016, clauses 7.1.5.1.1 (measurement system analysis) and 7.1.5.2.1 (calibration and verification records).
• ISO 9001:2015, clause 7.1.5.2 (measurement traceability).
• ISO/IEC 17025:2017, general requirements for the competence of testing and calibration laboratories.