A coordinate measuring machine can generate a polished report, but the report is only as useful as the requirement it evaluates. In molded plastics, a reported value can change with datum setup, part restraint, conditioning, elapsed time, sampling strategy and calculation. Buyers need an evidence chain: functional need becomes a critical-to-quality characteristic, the controlled definition states it clearly, the inspection recreates the intended reference frame, and a stated decision rule determines what happens next.
This matters when the design office, molder, metrology provider and receiving site work in different countries. European, North American and Middle East buyers may use different conventions, units, languages and customer templates, but need one controlled interpretation. This guide assumes the buyer owns product requirements and acceptance authority while the supplier proposes a practical measurement plan. It does not establish product compliance, certification, regulatory approval or a universal sampling requirement. Buyers should provide and verify applicable customer, sector and destination-market obligations before release.
Do not ask a CMM report to resolve an unclear drawing. First agree what feature matters, how the part is located, what condition it is measured in and who can accept a departure.
01
Build one requirement-to-decision chain before selecting instruments
Start with assembly or use, not the inspection room. Identify interfaces that could prevent fit, sealing, retention, clearance, movement, appearance or another agreed function. Then identify the characteristic that represents that risk. A hole pattern may locate a mating housing; a surface profile may support a gasket; a snap may govern retention. This prevents tight-looking dimensions from being mistaken for risks.
For each selected characteristic, record the controlled CAD and drawing revision, functional reason, requirement, datums, part condition, method, unit, sample identity and disposition owner. Call a feature an injection molding CTQ only when variation could affect an agreed outcome. Process settings, cosmetic preferences and reference dimensions may need control without becoming CTQs. An assembly check may be decisive even where a local dimension is measured.
The buyer fixes product intent, critical interfaces, acceptance criteria and any required report format. The molder or metrology provider proposes the fixture, point density, sequence and routine method. Review the proposal against a sample or representative fixture for complex geometry. This exposes assumptions instead of embedding them in a programme purchasing cannot interpret.
| Link in the chain | Buyer should define or confirm | Evidence should show |
|---|---|---|
| Functional requirement | What must assemble, seal, locate, move or look acceptable, and under which stated conditions. | The relationship between the characteristic and the intended use, not only a number on a report. |
| Drawing requirement | Controlled revision, datum references, units, applicable convention and acceptance criterion. | That the evaluated feature and reference frame match the released definition. |
| Measurement plan | Required evidence, part condition, sample or cavity identification, and any customer format. | Fixture, alignment, method, programme revision and results that can be interpreted later. |
| Release decision | Who accepts conforming evidence, concessions and corrective actions. | A clear pass, hold, concession or next action rather than an unqualified data file. |
02
Use GD&T to preserve function, not to decorate the drawing
Injection molding GD&T is useful when it communicates how a feature must relate to the part in service. A datum is not simply a surface for a probe; it is the reference that constrains the part to reflect function. Where a housing seats on pads and locates from a face and two features, discuss the inspection setup against that interface. A coordinate result that cannot reproduce the functional reference may not answer the fit question.
Review every geometric callout with its datum scheme, material condition where applicable, accessibility and likely part deformation. Name the governing convention in the purchase specification. ASME Y14.5 provides a common language for symbols, rules and interpretation; ISO GPS documents may be specified in other programmes. Do not mix conventions, model annotations and informal practice without declaring which interpretation controls. For a translated drawing, identify the controlled language and version hierarchy.
A molded part may flex, have textured or freeform surfaces, or lack stable datum features. The plan may therefore need a fixture, assembly-like restraint, profile evaluation to nominal CAD or a functional gauge. Request the rationale, not only a claim of measurement difficulty. The injection molding inspection plan should state free or restrained condition, datum contacts, and whether the result is geometric, size-based or functional.
03
Assign the CMM to the questions it can answer, and document its limits
Injection molding CMM inspection suits complex locations, orientations, profiles and multi-feature relationships when fixturing is repeatable and the programme reflects the drawing. It is not automatically the best routine check. An attribute gauge may be more direct for an assembly condition, while a caliper, optical system or functional fixture may suit another feature. Require an agreed method for critical evidence, rather than a CMM merely because the part is important.
For CMM work, request the programme revision, datum alignment and fixture, sensor approach, sampled features or areas, relevant calculation rules, units, part condition and sample, cavity or lot traceability. NIST notes that probe behaviour and sampling strategy affect CMM uncertainty. A calibrated machine is necessary but does not by itself show that the task-specific method is reliable.
A measurement-system study should fit the feature, tolerance, part behaviour, operator interaction and decision risk. Ask whether repeated use of the agreed method can support acceptance. For results near an agreed boundary, predefine treatment of uncertainty, rounding, remeasurement, independent review and any guard-banding. Do not invent the rule after a disputed report arrives.
| Evidence approach | Often useful when | Buyer question to ask |
|---|---|---|
| CMM programme | Several datums, spatial relationships, profiles or complex feature locations need coordinated evidence. | Does the alignment, fixture and sampled geometry represent the drawing and intended function? |
| Dedicated or functional gauge | The assembly condition, insertion, seating or pass-fail interface is the real product requirement. | Does the gauge reproduce the relevant mating condition, and who controls its acceptance basis? |
| Simple variable check | An accessible feature is being monitored efficiently after the acceptance method is established. | Is the method adequate for the required resolution, access and traceability? |
04
Make part state and time part of the specification
A molded part is not a fixed artifact. Resin grade, fillers, flow direction, wall transitions, cooling history, residual stress, humidity response and time after molding can affect a reading. Injection molding dimensional stability is therefore a planned condition of evaluation, not a promise implied by an early sample. The required conditioning and measurement window depend on material, geometry, use environment and customer requirements; this guide does not set them for a specific part.
For each critical evaluation, define or have the supplier propose the time reference, relevant storage or conditioning, temperature environment, fixture state, and free or restrained condition. Record the condition actually used. Label a learning trial as such rather than comparing it silently with a production acceptance result. Retain cavity identity where relevant, and include representative mating parts or a functional check when assembly is more sensitive than an isolated feature.
Early samples can reveal geometry but do not automatically demonstrate performance after an unreviewed material, tool, process or measurement-method change. Name changes that require impact review, new evidence or buyer approval.
05
Turn the plan into a cross-border acceptance protocol
Cross-border sourcing is stronger when the inspection plan is a controlled agreement, not a late attachment. State the part number, drawing and CAD revisions, tool or cavity configuration, material identity basis, units, drawing convention, language hierarchy, evidence package and acceptance authority. Use one primary unit system or state conversion and rounding rules. Every report and criterion should use the same declared basis.
At release, request a concise package linking sample identity to the approved revision and conditions: inspection-plan revision, actual results, method and fixture reference, agreed material and condition information, nonconformances, concessions and open actions. Request raw data, programme output, photographs, gauge records or independent-lab results only where they support a stated decision.
The injection molding quality control section should define routine monitoring, expected records or trends, escalation, containment, notification triggers and change authority. Typical triggers include drawing revisions, tooling modifications, cavity repair or replacement, relevant material or colour changes, a changed method, relocation and recurring nonconformance. These are governance prompts, not evidence of certification. Independently verify any certificate, laboratory accreditation, customer requirement or regulatory obligation material to the programme.
Practical takeaway
Before you move
to the next review.
- Create a requirement-to-decision register for each CTQ that names the functional reason, controlled revision, datums, part condition, method, record and disposition owner.
- Review GD&T with the proposed fixture and assembly context so datum simulation reflects the intended functional reference, not only convenient probe access.
- Approve the CMM programme or alternative method at the level needed for risk, including alignment, sampling, units, traceability and treatment of results near an acceptance boundary.
- State the conditioning, timing, free-state or restrained-state and cavity-identification rules needed to make dimensional evidence comparable.
- Put units, language hierarchy, evidence package, change-notification triggers, concessions and buyer verification duties into the purchase specification before release.
FAQ
Common questions
Should every CTQ be measured on a CMM?
No. A CTQ needs evidence that directly supports its functional acceptance decision. A CMM can be appropriate for complex geometric relationships, but a functional fixture, dedicated gauge or simple variable method may be more meaningful for a particular feature. Agree the method, part setup and report content before samples are evaluated.
What should a buyer do when a CMM result and an assembly result disagree?
First confirm that both assessments use the same controlled revision and that the measured part, conditioning state and fixtures are identified. Then review datum simulation, restraint, sampling strategy, calculation, mating-part condition and the functional requirement. Do not choose the preferred result by default. A documented correlation review may show that the drawing, inspection method or assembly check needs clarification.
Can a calibration certificate or a supplier quality certificate replace a part-specific plan?
No. Calibration and quality-system certificates can be useful supporting evidence, but they do not define the part’s datums, measurement strategy, acceptance rule or functional evidence. Verify any claimed certificate independently for the relevant organisation, site, scope and current status, then retain a part-specific plan for the actual release decision.
References
Source material
- [1] ASME, Y14.5 Dimensioning and Tolerancing overview.
- [2] National Institute of Standards and Technology, Measurement Uncertainty Considerations for Coordinate Measuring Machines, NISTIR 5170.
- [3] Automotive Industry Action Group, Quality Core Tools overview, including Control Plan and Measurement System Analysis.
- [4] International Organization for Standardization, ISO 286-1:2010 geometrical product specifications overview for linear-size tolerances and fits.
This guide is general information for early procurement and engineering discussion. It does not replace product-specific DFM, qualification, validation, regulatory assessment or legal advice.