Board Level Reliability Test Plan and Sample Size

A reliability test answers a specific question, and the question has to be written before the test is chosen. A plan that lists tests without stating what each one decides produces data that cannot be used to make a decision.

Starting From the Question

The questions are usually about a mechanism: whether a joint survives power cycling, whether a coating keeps moisture out, whether a connector survives mating. Each maps to a different test.

A test that does not map to a question is an expense. Our thermal cycling notes describe one of the mappings.

Choosing the Stress

The stress should reproduce the mechanism rather than the environment. A damp heat test reproduces humidity damage, and a thermal shock test reproduces the mechanical strain of a fast change, and they are not interchangeable.

Using the wrong stress produces a pass that means nothing. Our failure investigation notes describe how the mechanism is identified.

Sample Size and Confidence

The sample size follows from the failure rate that has to be detected and the confidence required. A test with three samples can show that something works and it cannot show that the failure rate is low.

The sample size should be stated with the result. Our yield analysis notes describe the statistics in the production context.

Failure Criteria

The criterion should be measurable: a resistance rise, a leak rate, a change in a functional parameter. A criterion based on a visible crack gives a different answer from one based on resistance.

The criterion has to be fixed before the test starts. Our joint criteria notes describe the measurement that supports it.

Acceleration and Its Assumptions

Acceleration is a model, and the model assumes a mechanism. Where the stress is increased far enough, the mechanism changes and the model stops applying.

The assumption should be written into the plan. Our fatigue notes describe the extrapolation in this case.

Correlating with the Field

A test result becomes useful when it is compared with returns from the field. The comparison requires the field conditions to be known, which means knowing the duty cycle of the product.

Without that, the acceleration factor has no anchor. Our field failure notes describe the data that is needed.

Verification

The verification is a written question for each test, a stress chosen to reproduce the mechanism, a sample size stated with the result, a failure criterion fixed in advance and a field comparison that states the assumed duty cycle.

Our quality notes describe how the records are kept.

Process Control and Verification

Keeping a sample from the panel turns a dispute into a measurement, because the same coupon can be re-examined by both parties without rebuilding the batch. Reviewing the design before the data is released is cheaper than correcting it after the panel is in the tank, because every step downstream inherits the decision made at the front end.

Documenting the assumption is part of the design work, and a short note on the drawing prevents a question that would otherwise arrive a day later and cost a day of schedule. The process window is set by the narrowest step in the flow, so an improvement anywhere else shows up as margin rather than as yield until that step is addressed.

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.

Where the process window is narrow, the measurement resolution has to be better than the window, or the data cannot distinguish a good part from a marginal one. A record that identifies the operator, the date and the settings is worth more than a record that identifies only the result.

The acceptance criteria should be written before the work starts, so that the decision is made by the specification rather than by the person inspecting.

FAQ

How many samples are enough? It depends on the failure rate to be demonstrated and the confidence required. The number should be calculated rather than chosen for convenience.

Can one test cover several mechanisms? Rarely well. A combined test is a compromise, and a failure in it is harder to attribute.

What does gopcb provide in a reliability plan? We provide a written question for each test, a stress chosen to reproduce the mechanism, a sample size stated with the result, a failure criterion fixed before the test starts, and a field comparison that states the duty cycle it assumes.