Plastic parts inspection and metrology should begin before a mold is released, not after parts arrive. A workable inspection plan connects every critical characteristic to a datum scheme, measurement method, conditioning requirement, sampling frequency, acceptance rule, and traceable record. This is especially important for molded plastics because temperature, moisture, fixturing force, shrinkage, and time after molding can change the measured result.
How to Build a Plastic Part Measurement Strategy
Start by classifying the drawing characteristics. Safety, sealing, mating, electrical spacing, bearing fits, and assembly interfaces normally require more control than nonfunctional cosmetic geometry. The drawing should identify the primary, secondary, and tertiary datums that represent how the part is located in its real assembly. The inspection method must then reproduce that constraint without flattening or distorting the component.
A useful plan answers six questions: what is measured, where it is measured, how it is held, when after molding it is measured, what instrument is used, and what record proves acceptance. A nominal value and tolerance alone are not enough. For example, a thin molded cover can pass when clamped to a CMM fixture but fail to mate in a free state. The inspection instruction must state whether the profile is evaluated restrained or unrestrained.
Datum Strategy and Measurement Conditioning
Datums should represent functional assembly surfaces rather than convenient toolroom surfaces. A stable three-point primary support, two-point secondary location, and one-point tertiary stop can establish repeatable orientation without over-constraining the part. Small bosses, flexible clips, gate vestiges, ejector locations, and textured surfaces are usually poor datum candidates unless they are functional.
Conditioning is part of the measurement system. Hygroscopic materials such as PA6 and PA66 can change dimensions as moisture content changes. Parts measured hot from the press may also continue to contract. The control plan should specify the time after molding, temperature, humidity or dry-as-molded condition, and any required stabilization. Supplier and customer must use the same condition when comparing results.
CMM vs Optical Inspection vs Gauges
| 방법 | Best Use | 주요 한계 | Required Evidence |
|---|---|---|---|
| CMM | Datum-based 3D geometry, hole position, profiles, and complex relationships | Probe force, fixturing, and sparse point strategy can misrepresent flexible surfaces | Program revision, fixture definition, probe qualification, and result report |
| Optical or vision system | Edges, small features, thin sections, pitch, and noncontact measurement | Edge recognition depends on contrast, focus, burrs, and translucency | Lighting setup, magnification, calibration, and measurement routine |
| Pin, plug, thread, or snap gauge | Fast attribute checks for holes, threads, and production limits | Shows pass/fail but not process trend or exact deviation | Gauge ID, calibration status, wear check, and acceptance rule |
| Functional fixture | Assembly fit, connector engagement, clip function, or multi-feature interfaces | May hide which individual feature caused a failure | Fixture drawing, master approval, maintenance plan, and correlation study |
| Surface or height instrument | Simple heights, thicknesses, flatness checks, and accessible dimensions | Operator technique and contact pressure can dominate the result | Work instruction, contact method, resolution, and repeatability study |
Method Selection by Feature
| 주요 기능 | Preferred Production Method | What the Record Should Show |
|---|---|---|
| Hole size and position | Pin gauge for limit check; CMM or vision for actual position | Actual result by cavity plus gauge or program ID |
| Thread or insert | Functional thread gauge, torque and pull-out test as required | Engagement result, torque/pull value and lot traceability |
| Sealing profile | CMM/profile scan plus approved leak or pressure test | Datum alignment, profile map and functional test result |
| Overall warpage | Defined free-state fixture, CMM, height gauge or optical scan | Support points, restraint condition and maximum deviation |
| Cosmetic texture | Approved limit samples under controlled lighting | Defect zone, viewing condition, inspector and disposition |
Holes, Pins and Threads
Use calibrated pins for production go/no-go decisions and a CMM or vision system when actual size and position are needed for capability analysis. Molded threads should be checked for functional engagement, start condition, flash, and torque performance. A thread gauge alone does not prove boss strength or insert pull-out performance.
Profiles, Sealing Lands and Warpage
Profile tolerances require a declared datum alignment and point distribution. Sealing lands often need both geometry and surface-continuity checks. Warpage should be measured in the same free-state or restrained condition specified on the drawing. Record the fixture load when restraint is allowed.
Texture and Cosmetic Surfaces
Texture, gloss, color, weld lines, sink, and flow marks need approved visual standards, controlled lighting, viewing distance, and defect zones. Cosmetic inspection should not be mixed with dimensional acceptance unless the requirements are clearly separated.
MSA, GR&R and Fixture Repeatability
A calibrated instrument can still produce unreliable data. Measurement system analysis evaluates the entire method: equipment, fixture, program, operator, environment, and part handling. Gauge resolution should normally be substantially finer than the tolerance being judged; a common planning target is 10:1, while 4:1 may be accepted for difficult applications with customer agreement.
For variable data, conduct a gauge repeatability and reproducibility study using representative parts across the process range. For attribute gauges, verify agreement against known conforming and nonconforming masters. Flexible parts deserve a separate fixture repeatability study because small changes in clamping sequence can move the result more than the manufacturing process itself.
Sampling Plans, Control Plans and SPC Handoff
First article inspection confirms that the released tool and process can produce the drawing. Routine control is different. Critical dimensions may require every-cavity checks at setup, periodic sampling by cavity, automated monitoring, or 100% functional gauging. Lower-risk dimensions can use reduced sampling after capability is demonstrated.
The control plan should list characteristic number, specification, method, sample size, frequency, reaction plan, and record location. SPC is useful only when the measurement system is capable and the process stream is separated correctly. Combining multiple cavities into one chart can conceal a cavity-specific shift.
Reports, Traceability and Nonconformance
A dimensional report should identify the part number and revision, tool and cavity, resin grade and lot, production date, process status, sample quantity, measurement method, instrument ID, result, tolerance, and disposition. Photos or screenshots are useful for complex datum alignments and profile results. Electronic records should retain the raw data or enough detail to reproduce the reported value.
When a result is outside tolerance, quarantine the affected lot, confirm the measurement method, determine the cavity and time window, and document correction. A deviation or concession must state the exact characteristic, quantity, duration, risk review, and approval authority. Do not silently adjust the CMM alignment or fixture to make a nonconforming part pass.
Plastic Part Inspection Checklist for Supplier Audits and RFQs
- Provide the controlled drawing, 3D model, revision, ballooned characteristics, and datum definitions.
- Identify critical, safety, sealing, mating, and cosmetic requirements.
- Define conditioning time, measurement temperature, moisture condition, and restrained/free-state rules.
- Agree on instruments, gauge resolution, fixtures, point strategy, and functional masters.
- Specify FAI sample quantity, cavity coverage, routine sampling, and capability expectations.
- Request calibration status, MSA or GR&R evidence, report format, traceability, and retention period.
- Define the nonconformance, deviation, corrective-action, and resubmission workflow.
For a supplier review, prepare the ballooned drawing, CAD model, resin requirement, critical characteristics, sample quantity, and required reporting template. Ask the supplier to confirm how the molding process, inspection method, and evidence package will be aligned before production release.
자주 묻는 질문
Can a CMM measure flexible plastic parts reliably?
Yes, when the datum scheme, fixture supports, probe force, alignment, and restrained or free-state condition are defined. Without those controls, the fixture may create the measured shape.
When is a functional gauge better than a CMM?
A functional gauge is often better for fast production checks when several features jointly control fit or assembly. CMM data remains useful for diagnosis, capability analysis, and first article reporting.
How should molded parts be conditioned before inspection?
Specify time after molding, temperature, humidity or dry condition, and any moisture conditioning. Both supplier and customer should measure under the same agreed state.
What is an acceptable gauge resolution?
A 10:1 tolerance-to-resolution ratio is a common planning target. A lower ratio may be workable when supported by MSA and customer agreement, especially for flexible or difficult geometry.
What inspection records should a supplier provide?
Request the ballooned report, part and drawing revision, cavity and resin-lot traceability, instrument IDs, actual results, calibration status, deviations, and approval records appropriate to the project.
Related Engineering Resources
Use the first article inspection and PPAP guide to plan sample approval workflows, and review injection molding tolerance stack-up when several molded features determine final assembly fit.
