Microsection Sample Preparation: 7 Steps to a Trustworthy Cross Section

A microsection is a polished slice through a plated hole or a pad, examined under a microscope to show plating thickness, barrel integrity and the interface between copper and laminate. It is the reference method behind most acceptance decisions on a PCB, and it is only as good as the sample preparation that produced it.

Poor preparation is deceptive rather than merely untidy. Rounded edges, smeared copper and pull-out along the hole wall all look like genuine defects, which means a bad lab result can reject a good board or, worse, hide a real problem behind an artefact that everyone dismisses.

What a Cross Section Is Used For

The cross section answers questions that no surface method can. It shows the thickness of the copper in the barrel and on the surface, the shape of the plating where it meets the inner layer, the fill of a via and the presence of voids, cracks or separation between the resin and the copper.

It also supports the process record. Acceptance limits for plating, for the hole wall and for thermal reliability are all written against what a section shows, so the section is the evidence that closes out a qualification or an investigation.

Sampling: Choosing the Coupon and the Location

The sample has to come from a place that represents the manufacturing process. A coupon that travelled with the panel is normally preferred, but where production boards are sectioned, the chosen hole should be one from the smallest size group and from a region where plating is expected to be thinnest.

Taking several holes from different positions shows the distribution rather than a single point. Judging a plating process on one favourable hole is one of the most common ways a marginal bath passes qualification, and the error is invisible if only one section is ever produced.

Potting, Mounting and Orientation

The sample is cut oversize, cleaned, and cast in a mounting resin under vacuum so that the resin fills the holes. Vacuum impregnation is what prevents the barrel from tearing during grinding, because an unfilled hole has nothing to support the plating as the abrasive passes over it.

Orientation decides what the section will reveal. Cutting through the centre of the hole gives a true measure of barrel thickness, while an off-centre cut gives an apparent thickness that is always greater. Marking the intended plane on the sample before cutting keeps the result interpretable.

Grinding: Removing Damage Without Rounding

Grinding removes the bulk of the material and establishes a flat plane. The risk is edge rounding, which erodes the copper at the surface of the sample and makes the outer plating look thinner than it is. Adequate mounting, controlled pressure and a sequence of progressively finer abrasives all limit it.

Heat is the other enemy. Grinding generates heat, and heat smears soft copper across the surface, filling voids and creating a false appearance of a sound joint. Water cooling, moderate speeds and short contact times keep the sample below the temperature where that happens.

Polishing and the Etch Before Examination

Polishing refines the surface to a mirror finish so that the boundary between copper, intermetallic and resin is visible. Particles left from a previous abrasive step are pushed into softer materials and appear later as bright specks, so the sample and the equipment both have to be cleaned between steps.

A light etch is often used to reveal grain structure and to separate layers that polish to a similar brightness. Etching is a judgement call, because too much etch removes the very plating that was to be measured and opens the boundary between layers that were genuinely joined.

Measuring Plating Thickness and Hole Wall Quality

Thickness is measured at defined points, usually at the barrel mid-wall and at the surface, using a calibrated eyepiece or image analysis. The measurement should record the position, because a barrel that is thick at the top and thin in the middle is a different defect from one that is thin everywhere.

The plating thickness figure alone rarely explains a failure. The same section shows whether the copper is uniform, whether it has cracked at the knee where the barrel meets the surface pad, and whether the inner layer connection is intact, all of which matter more than an average number.

Common Preparation Artefacts and How They Mislead

Rounding makes surface copper look thin, pull-out makes a bonded barrel look separated, smearing fills voids, and over-etching opens boundaries that were never open. Each of these artefacts mimics a defect that would cause a real rejection, and each is produced by the preparation rather than by the board.

The distinguishing feature is usually the pattern. Artefacts follow the polishing direction and appear on the outer surfaces, while genuine defects such as cracks and CAF damage tend to follow the material and appear inside the barrel or along a fibre path.

Photographing and Recording the Result

The image is the permanent record, so it should include a scale and enough context to identify the feature. Magnification, etch conditions, coupon identification and the measurement method all belong in the record, otherwise the image cannot be compared with a section taken six months earlier.

Consistent magnification between samples is what makes a comparison valid. A barrel photographed at a different magnification or with different illumination can look markedly better or worse than the one beside it in the file, and the change is easy to miss.

When the Section Disagrees With the Process Data

Disagreement between the section and the in-process measurements usually means the sample was not representative, or that the preparation has altered what it was meant to show. Before accepting either result, check the sampling position, the mounting quality and the etch, then repeat the section on a second hole.

Where the discrepancy persists, the copper thickness measurement methods used in production deserve the same scrutiny as the laboratory result, because a section is the reference only when it has been prepared properly.

Points to Confirm at First Article

The sequence of operations is part of the specification, because a different order produces a different result from the same steps. 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.

FAQ

Can a microsection be prepared without potting? It can be attempted, but the results are unreliable. Without vacuum impregnation the plating is unsupported during grinding, so it folds, tears or pulls away from the resin, and the barrel appears defective on a board that was sound.

How many holes should be sectioned? Enough to see the distribution across the panel and across the hole size groups. A single favourable hole proves very little, and the number should follow the risk of the plating process rather than the convenience of the laboratory.

Does a microsection damage the board? It destroys the sample, so production panels are sectioned only when the value of the information justifies the loss. This is the main reason coupons are designed into the panel border, so the analysis can be done without scrapping saleable product.