Aging Test and Burn-in for Industrial PCBA

A board that passes its functional test on the day it is built has demonstrated that it works, not that it will continue to work. The distinction matters most for industrial, medical and automotive products, where the equipment is expected to run for years in an environment that cycles through temperature, load and humidity. An aging test is the attempt to compress that exposure into a few hours or days, so that a unit which is going to fail early fails while it is still in the factory rather than at the customer’s site.

What the Test Is Trying to Accelerate

The failures that an aging test is designed to expose are not random. They cluster around three mechanisms. The first is the manufacturing defect that is present but not yet manifest: a joint that is partially wetted, a barrel that is porous, a component that was damaged by moisture during reflow or a connector that is not fully seated. These are the classic early failures, and they appear in the first hours of operation if they are going to appear at all.

The second is the marginal component. A device whose parameters sit at the edge of its tolerance may function at room temperature and drift outside the specification when it is warm, and a product built around it may pass every test that is performed at ambient conditions.

The third is the thermal design. A unit whose internal temperature rise is higher than intended will run its components closer to their limits than the design assumed, and the effect is not visible until the unit has been operating long enough for the internal temperatures to stabilise.

How the Stress Is Applied

The most common arrangement for an electronics product is a burn-in at elevated temperature with the unit powered and exercising its functions. Powering the unit while it is hot is what makes the test effective, because the temperature rise inside a powered unit is part of what the design has to survive, and a unit that is merely stored in a warm room has not experienced it.

Thermal cycling adds a second mechanism. Raising and lowering the temperature repeatedly subjects the joints and the plated barrels to expansion and contraction at different rates, which is the condition under which a marginal joint cracks. A barrel that is filled but porous will survive a constant temperature and will not necessarily survive a few hundred cycles.

Humidity is the third stress, and it is the one that exposes residue and insulation problems. Moisture absorbed into a residue creates a leakage path, and moisture inside a plastic package that has absorbed it during storage can cause damage when the part is heated. Where the product will operate outdoors or in a wash-down environment, the humidity exposure belongs in the test.

How Long and at What Level

The duration and the level are chosen from the failure mechanism being targeted and from the volume of product being tested. A short burn-in at the rated maximum temperature catches the assembly defects, because those mechanisms are fast. A thermal cycling regime that targets joint fatigue needs many cycles and therefore takes longer, which is why it is normally applied to a sample rather than to every unit.

The sample size matters as much as the individual test. A single unit that survives a hundred cycles has demonstrated that one unit survived. A small sample drawn across batches, with the failures recorded and classified, establishes a rate, and a rate is what allows a decision to be made about whether the process is stable. Where the purpose of the test is to qualify a process rather than to screen production, the sample is drawn deliberately to include units from different shifts, different material lots and different machines.

Screening Versus Qualification

The two purposes are frequently confused, and they require different tests. Screening is applied to production units to remove the ones that would fail early, and it is justified when the cost of a field failure is high enough to pay for the test on every unit. Qualification is applied to a sample to establish that the design and the process can meet the requirement, and its purpose is evidence rather than removal.

A burn-in that is used for screening has to be short enough not to dominate the production cost and severe enough to precipitate the defects being targeted. A qualification programme has to be severe enough to represent the product’s life and documented well enough to support the claim being made. Applying one and describing it as the other is a common and expensive mistake, because a screening test that screens nothing appears in the cost of every unit without changing the field failure rate.

Failure Analysis Is the Point

An aging test that records only how many units failed has produced a number. An aging test that records how they failed has produced information. Each failure is examined to establish the mechanism, and the mechanism is traced back to the process, the material or the design decision that produced it.

The distribution of the failures is as informative as their number. Failures that cluster in the first hour point to assembly defects. Failures that appear late in the run at a steady rate point to a wear-out mechanism or to a marginal component. Failures that concentrate in one position on the board, or in one batch of material, identify the specific cause rather than the general one.

The output of the exercise is a change: a process parameter, a material specification, a design revision. An aging test that produces failures and no changes has been an expensive confirmation that the product still has the faults it had before the test. Our assembly operation runs these programmes alongside the PCBA testing service for products built through industrial PCBA, medical PCBA and automotive PCBA, with the SMT assembly lines producing the units and the records held under quality management.

What It Costs and What It Buys

An aging test consumes test time, floor space, electrical energy and the labour of setting up and monitoring the units, and it therefore has to be justified against the cost of the failures it prevents. For a low value consumer product the arithmetic rarely works, because the cost of a field failure is bounded by the price of the unit and the cost of the return. For an industrial controller, a medical device or an automotive module the arithmetic is entirely different, because a field failure carries the cost of the intervention, the downtime and the operational consequences rather than only the replacement.

What the test buys is not a guarantee but an improvement in the failure distribution. Every unit that fails during burn-in is a unit that would otherwise have failed in the customer’s hands, and the population that remains has had its early failure mechanism removed. That is what the term reliability means in practice: not the absence of failures but their movement away from the customer’s experience and towards the factory, where they can be found, analysed and corrected.

The second thing the test buys is knowledge. A programme that runs for a year and records the mechanisms of its failures produces a picture of which joints, which components and which process steps are marginal in that product. That picture is not available from any inspection, because the failures it describes are the ones that only appear when the product is operated.

FAQ

Should every unit be burned in? Only where the cost of a field failure justifies it. Screening is an economic decision as much as a technical one.

Why power the units during the test? Because the internal temperature rise of a working unit is part of the stress the product has to survive.

What is the most valuable output? The failure mechanism and the change it produces, not the pass rate alone.