Field Returns: Diagnosis and Failure Classification

A returned unit is the only evidence about how the product behaves in the field. Handling it as a repair job rather than as an investigation wastes the information it carries.

The Return Process Begins Before the Unit Arrives

The information accompanying the unit is as valuable as the unit. Where it failed, when it failed, how long it had been running, the conditions and the firmware version all narrow the diagnosis.

Without that information the investigation starts from nothing, and the conclusion is usually that the unit cannot be reproduced. Our industrial assembly notes describe the records that make this possible for control equipment.

The packaging and handling of the return matter as well, since damage in transit produces a unit whose failure has nothing to do with its service life.

Classification Before Repair

Every return should be classified before it is repaired, at least into categories such as confirmed failure, intermittent, no fault found and damage.

The classification is the input to the reliability analysis, and it is lost if the unit is repaired and returned to stock without the record.

Where a unit cannot be reproduced in the test, it should be logged as intermittent rather than as no fault found, because the two have different causes and different follow-up.

Diagnosis Sequence

The diagnosis begins with the reported symptom and the unit’s history, and continues with a functional test in the original configuration.

Where the unit passes, the test should be repeated at the temperature extremes and under the conditions reported, since a fault that appears only when cold or hot cannot be found at ambient.

Where the unit fails, the failure should be localised before any repair is attempted, because repairing without identifying the cause destroys the evidence.

Root Cause and the Difference From Repair

Repair restores the unit to working order. Root cause identifies why it failed and what should change.

Identifying the root cause may require a section of a joint, an analysis of a component, or a review of the production records for the lot. These are separate activities from the repair and should be recorded separately.

Where the root cause is a production defect, the production records for the same lot should be reviewed to see whether other units are affected. Our quality notes describe the lot records that support this.

Intermittent Failures

An intermittent failure is the hardest case, because it does not appear when the unit is examined. The causes are usually thermal, mechanical or timing related.

The approaches that work are environmental stress screening, a long soak at the temperature extremes, and a review of the assembly for the defects that produce intermittent behaviour, such as a marginal solder joint or a connector with a poor contact.

A unit that passes should not be returned as good without a screening step, because an intermittent unit returned to the field will fail again and the second failure will cost more.

Feedback Into the Design

The returns data should feed back into the design and production reviews, with the failure rate by cause and by lot.

A cause that appears repeatedly is a design change waiting to be made, and a cause that appears in one lot is a production issue for that lot.

Where the product is mature, the returns data is the only new information about reliability, and it should be collected and analysed even when the volume is small. Our solder defects notes describe the failure signatures that are identified from returns.

Practical Rules

Collect the history with the unit, classify before repairing, and record the root cause separately from the repair.

Screen intermittent units rather than returning them, and review the other units from the same lot where the cause is a production defect.

Report the returns data to the design review, because a failure that is analysed and not communicated has been wasted.

Process Control and Verification

On a design of this kind, classification is the item that decides how the rest of the board is arranged. 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. The measurements that matter are the repeatable ones: conductor width and spacing, annular ring, finished hole size, plating thickness and surface finish are all verifiable on a coupon that travels with the panel.

Running a first article through the same checks as the production panel confirms that the two agree, and that comparison is the cheapest form of process control available at prototype stage. 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.

Process Control and Verification

On a design of this kind, classification is the item that decides how the rest of the board is arranged. 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. The measurements that matter are the repeatable ones: conductor width and spacing, annular ring, finished hole size, plating thickness and surface finish are all verifiable on a coupon that travels with the panel.

Running a first article through the same checks as the production panel confirms that the two agree, and that comparison is the cheapest form of process control available at prototype stage. 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.

Process Control and Verification

On a design of this kind, classification is the item that decides how the rest of the board is arranged. 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. The measurements that matter are the repeatable ones: conductor width and spacing, annular ring, finished hole size, plating thickness and surface finish are all verifiable on a coupon that travels with the panel.

Process Control and Verification

On a design of this kind, classification is the item that decides how the rest of the board is arranged. 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. The measurements that matter are the repeatable ones: conductor width and spacing, annular ring, finished hole size, plating thickness and surface finish are all verifiable on a coupon that travels with the panel.

FAQ

Should every return be repaired? Not before it is classified and diagnosed. A repair performed first removes the evidence.

What is the most useful single piece of information? The conditions under which the failure occurred, which is what allows the fault to be reproduced.

What does gopcb provide for field returns? We provide failure analysis including microsections and component analysis, lot traceability to identify other affected units, environmental screening for intermittent faults, and a root cause report that feeds back into the design and production reviews.