PCB Plating Thickness: Copper, Nickel and Gold Explained

Plating thickness is the specification that decides whether a board survives assembly and service, and it is also the specification most often quoted as a nominal value that hides a wide tolerance. Understanding what is being plated, why the thickness matters and how it is measured removes most of the confusion.

What Gets Plated on a PCB

Plating appears in three places. Copper is plated into drilled holes to form the barrels that connect layers, and it is plated onto outer layer surfaces during the same process. Nickel and gold are deposited as a surface finish, either as an electroless nickel immersion gold stack or as an electroplated hard gold for contact surfaces.

Each of these has its own thickness specification and its own measurement method, and conflating them is a common source of disagreement between designer and fabricator.

Copper Plating in the Barrel

The barrel is what carries current between layers, and its thickness determines whether it survives thermal cycling. As the board expands during reflow, the barrel is stretched in the z-axis, and a thin or brittle deposit cracks. A typical requirement is around 20 to 25 microns of copper in the barrel wall, measured at the centre of the board where plating is thinnest.

Aspect ratio is the controlling variable. The ratio of board thickness to hole diameter determines how easily the plating chemistry reaches the middle of the hole, and beyond roughly eight to one, uniformity becomes difficult without special agitation and chemistry. Thick boards and small holes therefore need more process control rather than simply more plating time.

Surface Copper and Finished Thickness

Outer layer copper is plated at the same time as the barrel, which means the finished conductor thickness exceeds the starting foil. A design that begins with one ounce foil typically ends up closer to one and a half ounces after plating, and impedance calculations must use the finished value.

Where heavy copper is required, the plating time increases and the surface becomes less uniform, which affects both the etch result and the assembly land pattern. The uniformity of the plated surface is a separate specification from the average thickness.

Electroless Nickel and Immersion Gold

In an ENIG finish the nickel layer is the functional one. It provides the solderable surface and the diffusion barrier, and a specification of roughly three to six microns is typical. The gold layer that protects it is very thin, usually in the range of 0.05 to 0.1 micron, and it dissolves into the solder during reflow.

Thicker gold is not an improvement. Above roughly 0.15 micron the gold can embrittle the joint as it forms intermetallic compounds, so the specification should state a range rather than a minimum. Nickel thickness and quality control the reliability of the joint; the gold is protection, not the soldering surface.

<img src="https://www.gopcba.com/wp-content/uploads/2026/09/218-1.jpg" alt="X-ray fluorescence measurement of ENIG plating thickness” />

Hard Gold for Contact Surfaces

Contact fingers and connector edges use electroplated gold over nickel with a hardening agent, and their thickness is specified in terms of wear rather than solderability. Common values range from about 0.5 micron for moderate insertion counts to several microns for harsh duty.

The distinction from ENIG is important in layout. Hard gold does not wet with solder reliably, so a pad that is both a contact and a solder point needs a selective plating process or a separate finish on that area.

How Thickness Is Measured

X-ray fluorescence is the standard non-destructive method. It measures the mass of coating per unit area and converts to thickness using the assumed density of the deposit, which means two instruments calibrated differently can report slightly different values.

Cross sectioning provides a direct measurement and is used for first article verification and for process validation. For barrel copper it is the only method that shows the distribution through the hole rather than an average, and it also reveals voids and cracks that a thickness reading cannot detect.

Specification and Tolerance

A specification should state the minimum, the nominal and the method of measurement. Quoting only a nominal value leaves the acceptable range undefined, and quoting only a minimum allows a result far above the requirement that may affect other characteristics.

For hole copper, state the measurement location: at the barrel wall in the middle of the board, not at the surface where plating is thickest. For surface finishes, state the substrate layer thickness rather than the total stack, and specify the plating process where the difference between electroless and electroplated deposits matters.

Why Thickness Problems Appear Late

Insufficient barrel copper causes no immediate failure at electrical test. The board passes continuity and isolation, works on the bench and fails during thermal cycling or after months in the field. That delayed failure is why plating quality is verified by coupon testing and microsection rather than by functional test alone.

Where the failure mode is a cracked barrel, the investigation should cover plating thickness, deposit ductility and the expansion characteristics of the laminate, and the information in copper plating defect prevention provides the starting point. Where the design also uses filled vias, via filling practice sets the additional requirements.

Plating Thickness in Filled and Capped Vias

Via in pad and filled via processes add a further thickness requirement. After the hole is filled with a conductive or non-conductive paste, the pad is plated over to produce a flat surface for the component ball. The plated cap must be thick enough to survive reflow and mechanical handling without cracking, and the specification usually calls for a minimum copper thickness over the fill as well as a flatness tolerance.

The result is a stack of materials with different expansion behaviour, so the process is qualified by thermal cycling and cross section rather than by a thickness reading alone. Where the design uses these structures on fine pitch devices, the plating specification and the filling specification are reviewed together rather than independently.

Plating Thickness and Solderability

Plating thickness affects assembly as well as reliability. A finish that is too thin oxidises in storage and loses solderability before the board reaches the line, while one that is too thick can interfere with the formation of a reliable joint. The usable window is defined by the finish specification rather than chosen by the designer.

Intermetallic growth is the mechanism behind both limits. As solder wets a metallic finish, intermetallic compounds form at the interface and grow during reflow and subsequent thermal exposure. A thin deposit can be consumed entirely, leaving a joint that depends on the underlying layer, while the growth rate determines how many reflow cycles the finish tolerates, as discussed in lead-free versus leaded solder.

Storage conditions complete the picture. Humidity and sulfur exposure degrade a finish over months, which is why the plating specification and the packaging specification are decided together for boards that will not be assembled immediately.

FAQ

Is more copper in the hole always better? Up to a point. Adequate thickness prevents cracking, but an excessively thick or brittle deposit can be less ductile and crack more readily. Deposit quality and ductility matter as much as thickness.

How is ENIG thickness verified on a production board? X-ray fluorescence on a sample coupon from the panel, with cross sectioning used for the first article. The measurement must distinguish the gold from the nickel layer, which the instrument does by using different excitation energies.

Can a board be replated if copper is thin? Not practically, because plating after lamination and etching would deposit copper everywhere. The correction happens at the process level, in the plating line parameters, rather than on the finished board.