Selective Plating Methods in PCB Fabrication

Most plating is applied to a whole panel. Selective plating applies metal only where it is needed, which saves expensive material and avoids coating surfaces that must remain solderable. Four techniques are used in PCB work, and each suits a different geometry and production volume.

Why Plating Is Applied Selectively

Gold, palladium and rhodium are expensive, and a thin layer is needed only where a contact has to be wiped, inserted repeatedly or held at a stable low resistance. Solderable pads need something quite different, usually a finish that dissolves into the joint rather than resisting it.

Blanket plating the whole panel with hard gold would be wasteful and would also make the pads difficult to solder. Selective methods solve the problem by masking the areas that should not be plated, or by bringing the electrolyte only to the area that should.

Tab and Finger Plating

The classic case is an edge connector, where the contact fingers need a hard, wear resistant surface. The panel is covered with a resist so that only the fingers are exposed, and the sequence runs through cleaning, a mild etch or scrub to activate the copper, a nickel strike, then the contact metal.

Nickel is deposited first, typically 2.5 to 6 µm, and the hard gold layer on top is usually 0.5 to 1.5 µm, alloyed with a small amount of cobalt or nickel to raise its hardness. Rhodium and palladium alloys are used where wear resistance or a very low contact resistance is essential. Thickness is verified by X-ray fluorescence on the plated fingers rather than assumed from the plating time.

Through-Hole Plating and Conductive Ink

Through hole plating is a selective process in the sense that the deposit must be continuous inside the barrel and adequate at the knee, where the barrel meets the surface pad. The standard route passes through desmear, electroless copper and electrolytic plating, with rinsing and chemical control at every stage.

A practical alternative for prototypes uses a conductive ink, applied so that it coats the hole wall with an adherent, conductive film that is cured and then plated directly. The approach removes the desmear and back-etch steps that ordinary plating requires, which shortens the process and reduces the chemical handling, at the cost of a different set of process controls.

<img src="https://www.gopcba.com/wp-content/uploads/2026/08/iStock-1307839840-jpg.webp" alt="Brush plating tool applying metal to a local area of a circuit board” />

Reel to Reel Selective Plating

Components such as connector pins, leadframes and flexible circuit strips are manufactured as a continuous metal strip and then stamped into shape. Plating such parts one at a time would be uneconomical, so the strip is passed through a sequence of cells and plated in a continuous line.

Areas that must not be plated are covered with a resist before the strip enters the line; the exposed sections receive nickel, gold, silver, rhodium, palladium or an alloy such as tin-nickel. The process gives excellent thickness control because the strip geometry and the current distribution are constant, and it is the standard way that high volume connector parts are finished.

Brush Plating for Repair

Brush plating, also called tampon plating, brings the electrolyte to the workpiece on an absorbent pad wrapped around an inert anode such as graphite. Only the area under the pad is plated, and the rest of the board is untouched.

The technique is used in repair shops and assembly houses rather than in panel fabrication. A worn gold finger, a damaged contact on an expensive assembly or a localised repair can be restored without stripping the whole board. Thickness control is manual and the result is not equivalent to tank plating, so it is a repair method rather than a production process. Where a whole edge connector is being reworked, a controlled approach is preferable, and the same care described for manufacturable design guidelines applies to the plating specification.

Masking and Its Cost

Every selective process needs a mask. On a printed circuit panel the mask is a resist applied and imaged photographically, which adds artwork, imaging and stripping steps. On a reel of stamped parts it is a patterned resist applied in the line.

The cost of masking has to be set against the cost of the plated metal. Where the plated area is small, selective plating wins easily; where almost the whole panel needs the same treatment, blanket plating may be cheaper despite the extra metal. The trade should be evaluated per product rather than assumed.

Choosing a Method

For edge connectors on a rigid board, tab plating is the standard answer. For connector parts made in high volume from strip, reel to reel plating is the only economic route. For a prototype board that needs plated barrels without a full electroless line, a conductive ink process may be quicker. For a repair, brush plating.

Whichever route is used, the fabrication drawing should state the finish for each region separately, with the thickness range, the underlayer and the surface that must remain solderable. Ambiguity in that area is what produces boards with gold on the pads that should have been solderable. The interaction between finish and assembly is covered in the comparison of lead-free versus leaded solder.

Process Control on a Selective Line

Selective plating lives or dies on masking accuracy. If the resist lifts at the edge of a finger, the deposit spreads onto the solder mask or onto an adjacent pad; if the opening is undersized, the plated area is short of the specification. The mask artwork should therefore be checked against the connector footprint before the panel is imaged, not after plating.

Current distribution is the second control. Fingers at the edge of a panel plate differently from those in the middle, and the difference grows as the plated area becomes small relative to the panel. Coupons placed at the panel edges and centre show whether the deposit is uniform, and plating time is adjusted per product rather than inherited from the last job.

Finally, the underlayer has to be controlled as carefully as the surface. A nickel layer that is too thin allows copper to diffuse into the gold and the contact resistance rises over time; a layer that is too thick cracks when the finger is flexed. Recording both thicknesses for each lot, and keeping the records with the panel, is what makes a contact failure investigable years later. The quality characteristics that matter across the whole board are described in the notes on PCB design quality characteristics.

A final note on documentation: state the plated area as a drawing feature rather than as a note on the drill table, and define which side of the panel carries the mask opening. A clear mechanical definition removes the ambiguity that otherwise leads to a re-plated or scrapped lot.

FAQ

Why is nickel plated under hard gold? Copper diffuses into gold and softens the contact. Nickel acts as a barrier and gives the gold a hard, stable base that resists wear.

Can selective plating be applied to a single pad? Yes, using masking or brush plating, though the cost per feature rises sharply as the plated area shrinks. Small quantities are usually handled by masking a panel rather than plating individual pads.

How is plated thickness verified? X-ray fluorescence on the finished part or on a coupon, backed by cross sectioning where the underlayer thickness matters. Plating time alone is not a reliable measure on a production line.