SMT Changeover Verification and First Article

A changeover is the moment when a line that was producing one product correctly starts producing another. Most of the escapes that follow a changeover are not random; they come from a step that was skipped because the line looked ready, and a verification that is written down and short is the control that catches them.

What Has to Change

The program, the feeder setup, the stencil, the paste, the nozzles, the reflow profile and the conveyor width are the items that move. Each has a physical and a software half, and the two can disagree without anything looking wrong.

The disagreement is the risk. A feeder loaded with the right part in the wrong lane produces a board that looks correct until the first test, and the recovery costs more than the verification would have. Our floor zoning notes describe how the line is arranged so that the changeover can be done in a controlled way.

The Verification List

A short list is better than a long one. The items that earn their place are the program identity, the feeder lane to part mapping, the stencil identity and revision, the paste lot and its thaw state, the nozzle set and the profile selection.

Each item is verified by reading it rather than by remembering it. A program name on a screen and a stencil in a frame are two facts that have to be matched, and matching them is a two minute activity. Our placement capability notes describe how the machine side of that is quantified.

First Article and Its Role

The first board is inspected against the drawing before the line runs at rate. That inspection is the only point at which a changeover error can be caught cheaply, because after an hour the line has produced a stack of the same board.

The first article is therefore a full inspection rather than a spot check: placement, polarity, paste volume on the critical pads and the profile from the latest run. Our first pass yield notes describe how the result is recorded against the product.

Common Changeover Escapes

The recurring escapes are a feeder in the wrong lane, a reel that was spliced with a different part, a stencil from the previous revision, a profile that was left from the previous product and a paste that was still cold from storage.

Each of those has a corresponding check on the list, which is the reason the list is written rather than recalled. A changeover without a list is a changeover that relies on the operator having a good shift.

Verification of the Verifier

A person can complete a list without performing the checks. The countermeasure is to make each check produce a record that can be compared: a scan of the reel, a photograph of the stencil, a profile printout from the run rather than from the library.

The record also makes the investigation possible when an escape does occur. Without it, the changeover cannot be reconstructed and the same mistake recurs. Our board quality notes describe the standard the checks are made against.

Timing and the Pressure to Skip

Changeover time is the pressure that produces a skipped check, so reducing the changeover time is itself a quality measure. A line that is quick to change is a line where the list is affordable.

The quickest changeovers come from setup external to the machine: kits prepared in advance, feeders loaded and verified off line, and stencils staged with the product. That preparation moves the work out of the window where it is under pressure.

Records and Improvement

Recording the changeover time and the escapes that followed makes the pattern visible. Where escapes cluster on one product or one shift, the cause is usually in the preparation rather than in the verification.

Keeping the first article with the batch closes the loop, because a failure found later can be compared with the board that was approved. That comparison is what turns a changeover record into a control.

Process Control and Verification

A short note on the drawing about handling, storage or packaging is often worth more than an extra decimal place on a tolerance. Where a value sits close to a process limit, the drawing should say so, since the shop can then open the process window rather than working to a nominal figure that carries no tolerance.

A stack-up that is drawn rather than described removes most of the ambiguity from a quotation, and it lets the fabricator price the board against the dielectric and copper weights that will actually be used. Where the requirement is not written down, the shop supplies its own default, and the default is chosen for the process rather than for the design.

Checks Before Release

The checks that matter are the ones performed on the product rather than on a sample kept for the purpose, because a coupon that travels with the panel is the only evidence about that panel. Where a requirement can be measured, it should be measured at the point of manufacture and recorded against the board or the lot it applies to.

A parameter that is set once and never re verified drifts, and the drift is usually discovered by a defect rather than by the record. The tooling, the material and the profile form one system, and a change to any of them should be assessed against the other two before it is released.

FAQ

Is a first article required on every changeover? On a line where the product changes, yes. Where the line runs one product continuously, the periodic check replaces it.

Can the machine verify its own setup? It can verify that the program matches the product and it cannot verify that the feeder was loaded with the part it claims.

What does gopcb provide for changeover control? We provide a written verification list with a record for each item, setup prepared outside the machine window, first article inspection before the line runs at rate, changeover time and escape data recorded per product, and records that let a changeover be reconstructed after an escape.