The samples arrived and they fit. What now?

Checking the sample carefully is exactly the point of having one

A sample arrives, and someone measures it. Every dimension, the finish, the marking, the fit into the actual housing. That is diligence, and it is what the sample is for.

It also feels conclusive, because the part in your hand either works or it does not. The part in your hand is one part, and what you are buying is a process.

A sample proves a part. You need a process proved.

When a first piece is checked and signed off, two different questions are on the table and they get answered very unevenly:

Aerospace first article inspection and automotive part approval both ask for more than a good first piece, because those are separate claims. You do not need their paperwork to borrow the distinction.

So approve the sample — and you approved the wrong thing

If the sample fits, you have learned that. Spending the review re-confirming it adds nothing. The useful questions are about the difference between this piece and the run.

• Same tooling? Ask whether the dies that made the sample are the dies that will make production. Tooling wears, and on rolled threads that can move the pitch diameter before anything looks wrong (rolling versus cutting).
• Same finish route? A sample plated on a different line, or in a smaller batch, is not evidence about the production coating — and coating thickness eats thread tolerance (plating and thread tolerance).
• Same amount of attention? A first piece often gets watched more closely than a production part will. That is usually not deception, just how first pieces get made. Ask it as a question — what will be different when you run five thousand? — and a supplier can answer it concretely.
• Measured, or just fitted? “It went in” is not a measurement. If nobody recorded actual values for the characteristics that matter, you have no baseline to compare a future batch against.

Write down what you approved

An approval that exists only as “the samples were fine” gives you nothing to point at later. When a production batch is not fine, the argument is about what was agreed, and that argument needs a document.

This is the same discipline as item 0 in the enquiry: state which part you mean, precisely enough that two people cannot read it differently.

What sample approval is actually good for

It cannot certify the batch — that is a sampling problem, and sampling is a decision rule about lots rather than a guarantee about parts. What it can do is catch the problems that are cheapest to fix before the run starts:

• A specification that fights itself — a coating that does not fit the thread allowance, or a head that cannot be formed at that size (head forming limits).
• An assumption nobody stated — untoleranced dimensions with no cited source (the title block), or two references disagreeing about the same feature (worked example).
• Parts that conform and still will not assemble — tolerance stack-up.

After the run, any of those can turn into scrap, rework, a delay or an argument about what the specification meant. That is the value of the stage.

Common questions

The sample fits and works. Why is that not enough to approve production?

Because it tells you about that one piece. What you actually need to know is whether the process behind it will keep producing pieces like it — next month, at fifty times the quantity, after the dies have been changed. Those are different questions, and a sample can only answer the first. This is not a fastener-specific idea: aerospace first article inspection under AS9102 and automotive PPAP both ask for evidence beyond a good first piece, though their scopes differ and neither is only about repeatability.

So what should I actually check when samples arrive?

Check the things that will change between the sample and the run, not the things that obviously already work. Whether the tooling that made the sample is the tooling that will make production. Whether the finish was applied by the same process and line. Whether any step was done by hand or under closer supervision than production will get. And whether the dimensions you care about were measured and recorded, or just eyeballed because the part fitted.

Are samples really made differently from production parts?

Sometimes, and it is usually not deception — a first piece often gets more attention simply because someone is watching it come off the machine. That is worth asking about rather than assuming either way. The useful question is not "was this sample special" but "what will be different when you run five thousand", which a supplier can answer concretely: same dies, same plating line, same operator attention or not.

What should I write down when I approve a sample?

What you approved, precisely enough that someone can disagree with you later. That means the drawing or part number revision, the measured values for the characteristics that matter, the finish and its acceptance criterion, and the date. An approval that exists only as "the samples were fine" gives you nothing to point at when a production batch is not.

If sample approval cannot prove the batch, what does?

Nothing does, entirely — which is the subject of the previous article in this series. Lot inspection is sampling, and sampling is a decision rule about batches rather than a guarantee about parts. What sample approval can do is catch the problems that are cheapest to fix before the run: a specification that conflicts with itself, tooling that is not what you assumed, a finish that does not fit the thread. Those are worth catching there, because after the run they are scrap.

Ask us

When we send samples we will tell you what was different about making them, and what will change at production quantity, because that is the part you cannot see from the pieces in your hand. sales@tigerfasteners.com