Plating thickness measurement: XRF, coulometric and cross-section

What a plating thickness callout actually means

A drawing that says tin 3 micrometres minimum is stating a local minimum, not an average over the part. The plating bath deposits unevenly: current density is higher at edges and corners, lower in recesses, and almost absent inside a deep hole unless the bath is agitated and the anode reaches. A flat face can read 5 micrometres while an edge on the same part reads 8 and a thread root reads 2.

That is why a thickness report without positions is close to useless. The question is not whether the part is plated. The question is whether the thinnest zone that has to resist corrosion, carry current or survive a bend is above the minimum. The measurement plan should name the zones before the first part is measured, and the drawing should say whether the minimum applies to all surfaces or only to functional ones.

XRF: fast, non destructive, sensitive to setup

X-ray fluorescence measures the spectrum that the coating and the substrate emit when excited. Because each element has its own energies, the instrument can separate a tin layer from a copper substrate and report the coating thickness. It is quick, it does not damage the part and it can be used on a production sample without scrapping it.

The limits are practical. The measurement area is a spot, typically a few millimetres across, so a small thread or a narrow edge is hard to isolate. The substrate and any intermediate layer must be known, because the calibration is built on them; a nickel underplate that is not declared will bias a tin reading. Very thin layers, below roughly a tenth of a micrometre for some combinations, sit near the detection limit. Curved, rough or plated plastic surfaces need their own calibration, and a reading taken on a curved contact face is an approximation.

• Calibrate with a certified reference standard of the same coating and a similar substrate.
• Check the reading on the standard before and after the sample set.
• Record spot size, dwell time and the collimator used.
• Measure the same geometric zone that the drawing calls critical.
• Treat a reading on a curved or threaded surface as indicative unless the method has been validated for it.

Coulometric: dissolving a known area

The coulometric method clamps a small cell against the surface, fills it with an electrolyte and passes a current that dissolves the coating down to the substrate. The time to reach the substrate, at a known current and area, gives the thickness. It is accurate on small, defined areas and works well on coatings that XRF struggles with, such as very thin layers or coatings on a complex substrate.

The method is destructive. The cell leaves a small dissolved spot, and the part is normally scrapped. It also depends on the operator sealing the cell properly and on the electrolyte matching the coating. A leak at the cell edge exposes more area than assumed and the result reads high. Use it for a dispute, a new bath qualification or a periodic check against the XRF, not as a routine line gauge.

Cross-section: the reference method

A microsection cuts the part, mounts it in resin, grinds and polishes the cut face and examines it under a microscope. The coating appears as a band whose width can be measured against a calibrated scale. It shows what the other methods infer: the actual layer, the interface, any porosity, the underplate and the substrate.

It is the method of last resort and the most informative. It reveals a nickel underplate that nobody specified, a zinc rich layer that is not what the certificate claims, or a crack that runs through the coating at a bend. The cost is time and destruction, and the result depends on the cut position. A section through a flat face says nothing about a thread root unless the cut passes through the thread.

Choosing a method

The choice follows the question. For a routine lot check on a flat contact face, XRF is the right tool: fast, non destructive and precise enough when calibrated. For a thin coating, a small area or a coating on a difficult substrate, coulometric is often better. For a dispute, a new bath, a failure analysis or a claim about an undeclared layer, the cross-section is the reference.

Whichever method is used, the report should state it. A thickness of 4 micrometres by XRF and 4 micrometres by cross-section are not the same claim, and a customer who assumes one method may reject a result produced by the other. Name the standard, the instrument, the calibration standard and the measurement positions.

How often to measure

Thickness is a bath property as much as a part property. It drifts with bath chemistry, current density, rack position and part geometry. A stable bath with a proven rack can be checked once per lot or per shift. A new bath, a new rack, a change of supplier or a change of part geometry needs a first article and a tighter check until the process is shown to hold.

The sample should include the zones that are hardest to plate, not only the easiest. If the drawing calls a minimum on a thread or a bore, the sample must include a sectioned or destructively tested part from that zone. Measuring only the flat face and reporting a pass is a common way to ship a part that fails in service.

The record a buyer should keep

A useful thickness record carries the part number, the lot, the coating specification, the method, the instrument, the calibration standard and every reading with its position. It should also carry the acceptance rule: the minimum, the maximum if one applies, and whether the value is an average or a local minimum. Without that, a table of numbers cannot be judged.

Keep the reference standard certificate with the record. An XRF result is only as good as the standard it was set against, and a standard that is out of date or of the wrong coating makes every reading questionable. For a cross-section, keep the micrograph with the scale bar and the measured points marked, because the image is the evidence.