First Article Inspection for Zippers: A Practical Checklist

Symptom

First article inspection is often reduced to a signed form with three measurements and a photograph. The lot is then released, and two weeks later a functional problem appears that the first article could have caught: a top stop 1 mm out of position, a slider that locks only intermittently, a shade that matches the swatch under office light but not under D65.

The symptom of a weak first article is not a failed inspection; it is a passing inspection followed by a predictable defect. If the same defect classes keep escaping into bulk, the first article is measuring the wrong things or measuring them without the right conditions.

• Bulk defects that were not present in the approved sample.
• First article form completed in under 30 minutes for a complex product.
• No functional test, only dimensional measurement.
• Colour approved under office lighting rather than D65.
• No record of the machine, mould or operator used for the first article.

Root Causes

First article failures come from an incomplete checklist, from a sample that is not representative, and from approval given without authority. All three are process design issues, not inspector diligence issues.

• Checklist copied from a generic template. It measures overall length and width but not pitch, locking force, stop position or pull strength.
• Sample taken from the best cavity. A first article must include every cavity and every colour in the run, or it proves nothing about the rest.
• No functional testing. Dimensional compliance does not predict slider feel, lock reliability or chain split resistance.
• Uncontrolled environment. Colour and dimension judged without specified lighting and conditioning produce results that cannot be reproduced.
• Approval by the wrong function. A first article signed only by production lacks the authority to stop a launch when a marginal result appears.
• No link to the process record. Without machine, mould and parameter settings, the approved sample cannot be reproduced after a changeover.

Troubleshooting Sequence

Build the checklist from the failure history. Every defect that has escaped in the last twelve months should have a corresponding first article check, and every safety characteristic should be verified 100 percent on the sample, not by attribute.

• Confirm the sample covers every cavity, every colour and the full size range in the order.
• Condition the sample for 24 hours before dimensional measurement.
• Measure length, width, pitch and stop position against the drawing with calibrated instruments.
• Test locking force, slider pull force, top stop pull and bottom stop pull on at least five pieces.
• Assess colour against the approved swatch under D65 at 300 mm.
• Run a salt spray or adhesion check on plated hardware if the order requires it.
• Record machine number, mould number, parameter sheet version and operator, and attach the record to the approved sample.

Corrective Actions

When a first article fails, do not approve conditionally with a promise to monitor. Either correct the process and re-submit, or obtain a documented deviation signed by the customer. Conditional approval without a deviation is the mechanism by which a known defect becomes a bulk defect.

When a first article passes but bulk fails, audit the difference between the two runs: material lot, mould, machine, operator and environment. In most cases one of those five changed between the first article and the bulk run, and the checklist did not require it to be recorded.

Prevention

Make the first article a gate, not a formality. Production cannot start until the signed article is attached to the order, and any change of material lot, mould or machine requires a new article. That rule alone removes most of the gap between approved sample and bulk reality.

Donglong Zipper keeps the approved first article sample and its measurement record in a sealed bag with the order file, and the line supervisor compares the first off-tool piece of each shift against it. YCC buyers should review the first article record at the start of production rather than at the end, because a record reviewed after shipment can only document a problem, not prevent one.

Finally, keep the checklist alive. Add a check every time a new defect escapes, and remove a check only when data shows the defect class has been eliminated. A first article that never changes is a first article that has stopped learning from the plant.

Define the re-qualification triggers in writing, because the gap between an approved sample and a bulk run is almost always caused by an unrecorded change. A new tape lot, a new mould, a different machine, a revised parameter sheet or a change of plating bath should each require a new first article, and the requirement should sit on the production order rather than in a procedure manual that nobody reads at the line. When a change is unavoidable mid-order, the first off-tool piece after the change is treated as a new article and compared with the sealed sample before the line continues.

Retain the record with the order for the product lifetime. A first article that cannot be produced when a claim arrives after two seasons has no value, and the measurement record is often the only evidence of what the process was actually doing on the day the lot was made.