A moulded sample fits. Is it ready for production approval?

What did that one successful assembly establish?

It showed that this sample fitted this mating part with this assembly method. It did not establish retention in use, fit across the allowed mating-part variation, or consistency between cavities.

Rather than approving an undefined “good part”, connect evidence to the requirements production parts must meet. For a clip, locating dimensions, latch engagement, retention performance and visible surfaces may require different evaluations.

The illustration separates measured features from the area carrying load in the assembly. A complete dimension table does not by itself demonstrate retention. Equally, an acceptable assembly feel cannot replace a required dimensional inspection.

Measurements need a material and a measurement state

Start by linking the specified material to the material actually moulded: manufacturer, grade and lot. Using the same polymer family does not ensure the same grade, additives or conditioning history. Keep an unapproved alternative separate from evidence for the specified material.

Kościuszko and colleagues report time-dependent changes after moulding for one PP grade.[1] The practical implication here is to record the state at measurement, not to copy that study’s waiting periods into another product’s acceptance plan.

BASF discusses moisture conditioning for polyamides.[2] This is distinct from the PP study’s changes over time after moulding. Use the actual grade guidance and service conditions to agree storage, any required conditioning and measurement timing. Measuring at different states at the supplier and receiving site makes discrepancies harder to investigate.

Appearance and function answer different questions

For appearance, lighting, viewing distance, designated surfaces and agreed defect examples can make “looks good” more reproducible. Retention needs an evaluation linked to use, with the mating component, assembly method and relevant service or repeated-use conditions defined.

Myers and colleagues first identified a visually acceptable process window, then narrowed it using tensile-property results.[3] Appearance alone therefore did not settle the evaluation. The settings from that experiment are not a recipe for this clip.

For the example, the quality team would agree a retention evaluation and link tested sample IDs to their dimension results. A record showing evidence received and tests pending is more useful than a single line saying “all satisfactory”. It also explains which later steps must wait.

Keep the identity when you sample several parts

With a multi-cavity mould, retain the identity of the individual cavity. Mixing samples in one unlabelled bag removes the ability to trace a cavity-specific difference. Connect each sample ID to cavity, moulding conditions and shot or collection point.

Sample numbers and collection intervals follow the product risk and evaluation purpose. A handful of early pieces does not demonstrate long-term process stability. Checking initial geometry and function is related to, but different from, demonstrating a stable production process.

If a feature does not comply, record whether the next step is corrected samples or a specifically authorised concession with defined scope and duration. An empty result must not quietly become a pass. Name the pending evidence and the operation it prevents from proceeding.

Make approval useful for the next batch

Approval evidence should connect appearance, dimensions and function to one drawing revision and material specification. Identify and date retained reference samples, and keep the corresponding results. A physical sample alone cannot explain the conditions under which it was accepted.

Also decide what changes to material, tooling or processing require renewed evaluation. The aim is to move from “this one worked” to an intelligible basis for receiving future production parts.