Microsection Preparation and Interpretation

A microsection is the only way to see what is actually inside a board or a solder joint. X-ray shows density, electrical test shows continuity, and visual inspection shows the surface, but a polished cross section shows the copper thickness in a barrel, the intermetallic layer at a joint and the presence of a crack that has not yet opened. It is a destructive test, and its value depends entirely on how well the sample was prepared.

What a Microsection Shows

The technique reveals layer count and alignment, copper thickness on the surface and in the barrel, plating quality and adhesion, and the structure of a solder joint including its intermetallic layers and voids. It also shows defects such as resin recession, wicking, cracked barrels and delamination that no other method can confirm.

Because it is destructive, it is used on coupons, on process qualification samples and on failed units. The result applies to the specific location that was cut, which is why sample selection is as important as preparation.

Choosing and Cutting Samples

Samples should be taken from locations that represent the process. For a plated through hole evaluation, the sample should include holes of the smallest and largest diameter and should be taken from the panel positions that are most and least favoured by the plating cell. A single hole from the centre of a panel says very little.

Cutting must not damage what will be examined. An abrasive saw that heats the sample can smear copper or smear resin across the surface, and a rough cut close to the target leaves a damaged zone. The standard approach is a cut well away from the area of interest, followed by precision sectioning.

Potting and Mounting

The sample is encapsulated in a resin that supports the edges of the cut and prevents the copper and laminate from smearing during grinding. Vacuum impregnation fills voids and through holes so that the resin supports the features that matter, and it is essential when the sample contains barrels or a porous joint.

The choice of resin affects the result. A hard, brittle resin holds the copper well but can chip at a soft solder joint, while a softer resin protects the joint but allows more edge rounding. Mixed resin systems and edge retention additives exist to solve exactly this conflict.

Grinding and Polishing

Grinding removes the damage left by cutting, and polishing produces the surface that will be examined. The sequence uses progressively finer abrasives, with the sample rotated between steps so that scratches from the previous stage are removed rather than deepened. Each step must remove the damage from the step before it.

The most common preparation fault is over-polishing. It rounds the edges of the copper, makes a thin plating look thicker than it is, and can pull the intermetallic layer out of a joint. A properly prepared section has flat layers with sharp interfaces and a scratch pattern that is uniform across the surface.

Measurement and Interpretation

Measurements are taken with a calibrated microscope or an image analysis system. Copper thickness in a barrel is measured at several points, because plating is rarely uniform around the circumference, and the minimum value is the one that matters for reliability. The acceptance criteria come from the applicable standard rather than from the observer.

For a solder joint, the thickness of the intermetallic layer, the presence of voids and the shape of the fillet are the observations that matter. A layer that is too thick indicates excessive heat or time, while a layer that is incomplete indicates insufficient heat, and the two look similar to an untrained eye. The measurement detail that supports this sits alongside the plating controls in our note on plating thickness.

Common Artefacts

Not everything seen in a section is real. Smearing can close a genuine crack, edge rounding can hide a thin layer, and pull-out can remove a particle and leave a void that was not there. Preparation artefacts have characteristic shapes, and knowing them prevents a false conclusion.

The opposite error is dismissing a real defect as an artefact. The way to separate the two is to prepare a second sample from an adjacent location using a gentler sequence. If the feature appears in both, it is real.

Barrel Quality and Plated Through Holes

For plated through holes, the observations are copper thickness, uniform coverage around the circumference, the presence of plating nodules or voids and the corner where the barrel meets the surface pad. The corner is where thermal stress concentrates and where cracks begin, so it deserves attention in every sample.

Barrel quality depends on the drilling and plating process, and the parameters that control it are described in our article on hole copper. Microsections taken before and after thermal cycling show whether the plating has the ductility to survive assembly and service.

Reporting the Result

A useful report states the sample location, the preparation method, the measurements with their uncertainty and photographs at a defined magnification. Without the location, the result cannot be related to the process, and without the preparation details a second laboratory cannot reproduce it.

Photographs should include a scale and should show both the overall feature and the detail in question. The same discipline used to judge a finished board, described in our article on solder joint acceptance criteria, applies to the evidence presented in a section report.

Process Control and Verification

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.

FAQ

How many samples are needed? Enough to cover the process extremes: smallest and largest holes, and the panel positions with the least and most plating. A single sample is rarely sufficient for a qualification.

Can a microsection be repeated? Not on the same location. Grinding further removes material and reveals a plane behind the one examined, which is useful for a three dimensional defect but not for repeating a measurement.

What magnification should be used? Low magnification for the overview and higher magnification for layer measurement. Measuring a thin layer at low magnification is the usual source of an inaccurate result.