First article samples decide whether a program moves to production. They are also, too often, reviewed by holding the part next to a drawing and judging whether it looks right. Appearance tells you very little about a sintered metal part: the dimensions that matter are the ones inside bores, on datums and across fits.
This guide covers how to structure a first article review for MIM and PM parts — what to measure, what to ask for in writing, and how to handle the results that fall outside tolerance.
Start with the drawing, not the part
Before measuring anything, confirm which revision of the drawing the supplier built to. Samples that are dimensionally correct against an obsolete revision are a recurring and entirely avoidable source of confusion.
Then read the datum scheme. MIM and PM parts are measured from datums established after sintering, and the drawing should define them. A first article report that records dimensions without stating how the part was fixtured is not verifiable — you cannot tell whether a reading reflects the part or the setup.
Finally, identify the feature set that actually matters. Typically this is a short list: the fits, the bores, the sealing or bearing surfaces, the datums and any dimension whose deviation would cause an assembly or field failure. Everything else is secondary, and treating it as equally important wastes inspection budget and delays the review.
What to measure, what to leave
A first article report should cover every dimension on the drawing once, but the interpretation should be tiered.
| Tier | What it covers | Acceptance approach |
|---|---|---|
| Critical | Fits, bores, sealing faces, bearing journals, datums | Compare against drawing tolerance feature by feature; require a capability statement, not just a reading |
| Significant | Assembly clearances, mating dimensions, wall sections | Verify against tolerance with the process capability in mind |
| General | Cosmetic and non-functional dimensions | Confirm within general tolerance note; do not over-interpret single readings |
A single sample tells you whether the part is in tolerance. It does not tell you whether the process is capable of staying there. Where a critical feature sits mid-tolerance, that is reassuring. Where it sits on the tolerance limit, ask what the producer expects the variation to be across a run — a first article that only just passes is a forecast of rejects.
Check the material, not only the dimensions
Dimensional conformance says nothing about whether the part has the properties the application needs. Three checks cover most of the risk.
- Density. Usually measured by the Archimedes method and reported as a percentage of theoretical density. MIM normally reaches 95–99%; a figure well below the specification indicates a furnace or feedstock problem that will affect strength, sealing and magnetic response.
- Hardness. Confirms that the specified heat treatment was applied and produced the intended result. A hardness reading also flags an unintended condition — 17-4PH that was never aged, or a martensitic grade that was over-tempered.
- Chemical composition. Supplied by the powder producer and confirmed on the material certificate. This is the only practical way to detect a grade substitution, and it is worth checking when the application is sensitive to alloy content.
Density is the underrated first-article check. It is quick, it correlates with mechanical and magnetic performance, and it is the earliest indicator that the sintering process was not where it should be. If a part is dimensionally correct but low on density, the dimensional pass is not good news — it means the shrinkage was compensated for the wrong reason.
Review fit and function together
A dimension can be inside tolerance and still not work. This happens most often at fits and at features that interact with a mating part. Where the design permits, ask for samples to be checked against the mating component rather than measured in isolation — an assembly test catches form and position problems that a dimensional report, taken feature by feature, does not.
The same applies to appearance. If the part is visible to the end customer, inspect the finish under the lighting conditions the customer will see, not under a bench lamp. Cosmetic disagreements at first article review are common and are much cheaper to settle with a photograph and an agreed standard than with a rejected lot.
Reports worth requesting
- Dimensional first article report — every drawing dimension, with the drawing revision, measurement method and datum reference stated.
- Material certificate — alloy grade, heat or lot number, and chemical composition from the powder supplier.
- Density report — measured value and method, usually Archimedes.
- Hardness report — the scale used, the readings and the test locations.
- Process capability data where the program requires it — readings from a run, not from a single sample, which is what actually supports a capability claim.
- Inspection plan — what will be checked in production, how often, and with what method.
Where a formal submission is required, these documents form the backbone of a production part approval package. Building them into the first article review is easier than assembling them afterwards.
When results fall outside tolerance
This is normal, and it is more informative than a clean first article. What matters is the cause and the response.
- Establish whether the deviation is a measurement or a part problem. A dimension measured differently between the two parties is common, especially on features that are difficult to fixture. Confirm the method before accepting a deviation.
- Identify whether the cause is tooling, process or material. A dimensional offset across many features usually means shrinkage compensation; a deviation on one feature usually means a tool correction; a density or hardness miss means a process parameter.
- Ask what the correction will be and whether it requires tool work. Some corrections are process adjustments that cost a run; others require a mould change with a lead time.
- Get the corrected plan in writing, with a date. The point of a discrepancy report is to close it, not to record it.
Closing out from sample to production
Production approval is not the same as passing a first article. Before releasing a program, confirm four things: that the first article is approved against the current drawing revision, that density, hardness and material records are on file, that the production inspection plan matches the features that actually matter, and that both parties agree on the process window that will be held.
That last item is the one most often skipped. A first article produced under special attention is not evidence that routine production will match it. Agreement on the parameters that will be monitored — sintering profile, feedstock lot control, tool maintenance intervals — is what makes the first article a predictor rather than an isolated good result.
Conclusion
A first article review is a structured comparison between a part, a drawing revision and a specification. Dimensions are the visible part, but density, hardness and material records are what confirm the part is made of what it claims to be.
The most valuable output of the process is not an approval signature. It is a clear understanding of which features carry the function, how much margin they have, and what will be monitored in production. Programs that establish that at first article stage tend to run without surprises; programs that do not tend to rediscover it at a rejected lot.
Related component families
These pages cover the component families where first article requirements tend to be most demanding.
Precision Small Components
Small parts where measurement method matters as much as the tolerance.
Component guide →Component Design Notes
Feature-by-feature DFM notes that also define what to inspect.
Component guide →Custom OEM Components
Programs where approval documentation is required by the end customer.
Component guide →Quality System
How SINTS controls measurement, records and process consistency.
Quality overview →Have a drawing to discuss?
Send it to our engineering team. We review process fit and identify the information needed for a quotation.
Send Drawing for Review