Solder Joint Workmanship Standards and Acceptance Criteria

A solder joint is acceptable or it is not, and the decision has to be the same whoever makes it. A written workmanship standard is what makes that possible: it defines the appearance of a good joint, describes the defects that are not acceptable and gives the inspector a reference rather than an opinion.

The standard is also a design tool. A layout that cannot produce joints meeting the criteria is a layout problem, and knowing the criteria while the board is being designed is cheaper than discovering them at the inspection station.

Why a Written Standard Matters

Without a standard, acceptance depends on the experience of the inspector and on the pressure of the schedule. Two people looking at the same joint reach different conclusions, and the difference grows when the product is behind schedule.

The standard also protects the process. Where a defect is defined, the rate at which it occurs can be measured, and a measurement is what allows the process to be improved. An opinion cannot be trended.

Classes of Product and Their Requirements

Acceptance criteria are usually organised into classes that reflect the consequence of failure. A general purpose product, a product where continued performance is required and a product where failure could be hazardous each have their own limits on the same joint.

The class should be stated on the documentation, because an inspector who is not told which class applies will apply the strictest one, and the strictest one may not be achievable at the specified yield. The choice is a commercial and a technical decision that belongs to the product owner.

<img src="https://www.gopcba.com/wp-content/uploads/2026/08/pcb3.jpg" alt="Acceptable solder fillet on a through hole joint” />

What a Fillet Should Look Like

The fillet is the visible evidence that the solder wetted the pad and the lead. It should be concave, it should rise up the lead and spread onto the pad, and its surface should be smooth rather than granular.

The volume is judged from the shape rather than measured. A fillet that is too small indicates insufficient solder or a joint that did not reach temperature; one that is too large and convex indicates excess solder, which can hide a joint that never wetted underneath.

Wetting and Its Indicators

Wetting is shown by the contact angle between the solder and the metal. A small angle means the solder has spread and adhered; a large angle, or a solder that sits as a ball on the surface, means it has not.

The distinction matters because a joint can be mechanically sound and electrically continuous while showing poor wetting, and poor wetting is the indicator of a process problem that will produce worse joints later. That is why the criteria are written in terms of the appearance rather than the continuity.

Common Defects and How They Are Judged

Insufficient solder, excess solder, a disturbed joint, a bridge and a lifted lead are the defects that appear most often. Each has a characteristic appearance, and each has an acceptance limit that depends on where it is and on the class of the product.

A disturbed joint is the one that is most often missed. The surface is rough and dull, which indicates that the alloy was moved while it was solidifying. It is a defect even when the joint is complete, because the grain structure is weaker than a joint that cooled undisturbed.

Voids, Pinholes and Their Acceptance

A pinhole or a small void on the surface of a fillet is usually acceptable, provided the joint is otherwise sound. A void that occupies a significant part of the joint or that sits at the interface is a different matter, and it is judged by its effect on the mechanical and thermal path.

Where the void cannot be seen, as under an area array, the criteria have to be expressed in terms of the measurement rather than the appearance, which is why X-ray acceptance limits are written separately from the visual ones.

Inspection Aids and Magnification

The criteria are meaningless without a magnification to apply them at. A joint that appears acceptable at one power may show a defect at another, so the standard specifies the inspection criteria together with the magnification for visual inspection and the lighting conditions under which it is performed.

Aids such as a low power microscope, a magnifier lamp or an endoscope for a hidden joint are chosen for the feature being inspected. The choice should be recorded, because the same joint inspected with a different aid produces a different verdict.

Photographs as the Reference

A written description is open to interpretation; a photograph of the actual product is not, which is why the inspection criteria are usually issued with a picture set. Building a reference set from the line’s own joints, including both acceptable and rejected examples, is the most effective way to make the criteria understood.

The set should be updated when the product or the process changes, because the appearance of a good joint changes with the alloy and the finish. A reference set from a leaded process is misleading when the line has moved to a lead free alloy, which has a different surface appearance even when the joint is sound.

Applying the Standard on the Line

The standard has to be available at the point of inspection, not in an office. A laminated sheet at the station, or a photograph in the inspection software, means the inspector can compare directly rather than from memory.

The findings should be recorded by defect type, because the distribution is the diagnosis. A defect that appears only at one station points at the process there, and one that follows a component points at the material, which is the same reasoning used in any quality control programme.

Training and Consistency

Training against the standard is what makes it real. Two inspectors should look at the same set of joints and reach the same conclusion, and the exercise should be repeated when the standard is revised or a new product is introduced.

Consistency between inspectors can be measured with a set of reference samples and a simple comparison of the results. Where the disagreement is large, the criteria are ambiguous, which is a problem with the standard rather than with the people applying it, and it is worth fixing before the next audit.

Documentation and Revision Control

The standard is a document, and it has to be controlled like one. The revision in use should be identified at the inspection station, and the change from the previous revision should be communicated rather than left to be discovered.

Where a customer or an internal procedure defines the criteria, the assembly documentation should refer to that source as well, so that the inspector knows which document takes precedence if the two differ. That traceability is the same principle as the one applied to process documentation elsewhere in the build.

FAQ

Who decides the class of a product? The product owner, with the engineering and quality functions. It has a direct effect on the acceptable yield and on the cost of the assembly.

Can a defect be accepted by concession? Yes, where it is documented, justified and approved. An undocumented concession is simply an escape.

Do the criteria differ for hand soldering? The geometry differs because the process differs, but the principle of wetting and fillet shape is the same, and the same standard applies.