Gold Finger and Edge Connector Protection
A gold finger is a plated contact at the edge of a board that mates with a socket or a card-edge connector. It is the one part of the assembly that is expected to survive repeated insertion without any solder joint to hold it, and it is also the part most exposed to handling during the rest of the build. The plating is thin, measured in fractions of a micrometre of gold over a harder barrier, and the damage that matters is rarely visible as a scratch; it is a wear or contamination change that shows up as contact resistance much later.
What Damages a Gold Finger
The plating system is usually hard gold over nickel over the copper of the board edge. The nickel provides the hardness that resists wear, and the gold provides the low, stable contact resistance. Damage occurs when the nickel is exposed, when the gold is worn through, or when a foreign material is embedded in the surface. Scratching through the gold to the nickel is survivable if the exposed area is small; scratching through both layers to the copper starts corrosion immediately.
Contamination is the quieter problem. Solder paste, flux, adhesive, conformal coating and even skin oils can leave a film that increases contact resistance or that prevents the socket from making proper contact. Because the fingers are at the board edge, they are the first thing a hand touches when a board is picked up, and the last thing anyone thinks about when a process is set up.
Masking and Exposed Area Control
The plating pattern defines which areas receive the hard gold and which are protected. Solder mask should stop short of the contact area, and the transition should be positioned so that the contact zone is fully plated and the mask does not encroach on the mating length. Where the plating is done as a selective edge plating, the boundary between the gold and the tin or silver finish elsewhere on the edge has to be controlled, or the two finishes will interdiffuse at the boundary during reflow.
The exposed copper at the edge also needs protection through the process. A common arrangement is to plate the fingers with the final finish and then cover them with a masking tape or a peelable mask for the rest of the build, removing the protection as late as possible. Where a peelable mask is used, its adhesion and its removal residue both matter: a mask that leaves a film changes the contact resistance, and a mask that adheres too strongly can lift the plating when it is removed.
Bevel and Chamfer Requirements
An edge connector needs a bevel on the leading edge so that it enters the socket without lifting the contacts. The bevel is usually 20 to 30 degrees over a depth of about 0.5 to 1.0 mm, and it should be applied without chipping the laminate or smearing the plating. A bevel applied after plating can drag the gold over the edge and bury a burr of nickel and copper under it, which is a defect that inspection under magnification will find and a visual check will not.
The bevel angle also affects insertion force. A shallow bevel reduces the initial force but lengthens the ramp, while a steep bevel produces a sharp rise that can damage the socket contact springs. Where the product is inserted by an operator rather than by a machine, a slightly shallower bevel is often chosen for comfort. The requirement belongs on the fabrication drawing, with the angle, the depth and the finish condition all stated. The edge plating notes describe the process conditions that keep the bevel sound.
Handling Through Assembly and Test
The fingers should be protected from placement through soldering, and the protection should only be removed immediately before the board is packed. Where the protection is removed earlier, the fingers are exposed to the reflow oven, to the wash, and to every handling step that follows, and the probability of damage rises steeply. A peelable mask that survives the full process and removes cleanly is worth more than a cheap tape that has to be replaced twice.
Where the board is handled by the edge, gloves are not a substitute for protection. Nitrile gloves leave a film of plasticiser on the plating, which raises contact resistance; cotton gloves shed fibres that lodge in the socket. The correct practice is to handle the board by a non-contact area or by the protected fingers, and where the fingers must be touched, to clean them afterwards rather than assume the touch was harmless.
Cleaning and Chemical Compatibility
Gold is chemically stable, but the underlying layers are not. A wash chemistry that reaches a scratch, a plating pore or the boundary between the gold and the mask can attack the nickel and produce a corrosion product that grows along the interface. This is why a board with fingers should be washed before the tape is removed, and why a no-clean process is attractive where the residue can be tolerated.
Where cleaning is required after the fingers are exposed, the chemistry should be checked against the plating system. Alkaline cleaners attack some nickel layers, and acidic cleaners attack the nickel more readily when the gold has a pore. A simple test, immersing a sample of the same plated edge in the cleaner for the dwell time and inspecting before and after, answers the question and takes an hour. The surface finish guidance covers the chemistry of the layers, and the finish selection notes explain how the choice interacts with the rest of the assembly.
Insertion Cycles and Wear
The number of insertions a finger can survive depends on the hardness of the nickel, the thickness of the gold and the contact pressure of the socket. A hard gold of 0.8 micrometres over a hard nickel is often rated for hundreds of cycles, while a thin soft gold of 0.1 micrometre may be good for a dozen. The rating is a property of the system, not of the gold alone, and it should be obtained from the plating specification rather than assumed from the gold thickness.
Where the product is expected to be inserted and removed frequently, the contact resistance should be measured before and after a defined number of cycles on a sample. The change, rather than the absolute value, is the useful number, and a rise above a stated limit indicates that the plating or the socket is not suited to the application. The test also reveals whether the wear is adhesive, leaving gold on the socket contact, or abrasive, wearing the finger itself, because the two lead to different remedies.
Inspection Criteria for Plated Contacts
Inspection of the fingers is visual under magnification, with criteria written for scratches, for exposed nickel, for exposed copper and for contamination. A scratched area within the mating zone that exposes nickel is usually a reject; the same scratch outside the mating zone may be acceptable. The criteria should therefore be tied to the mating length, which is a drawing dimension, rather than to the whole finger.
Contact resistance is verified electrically on a sample, using a probe arrangement that reproduces the socket contact rather than a sharp probe that penetrates the surface. A sharp probe measures the resistance of the layer it has pierced, which is always low and tells nothing about the surface. Where the assembly has many fingers, a continuity test across the connector is a quick check, but the resistance measurement remains the only way to see a marginal contact. The four-wire method is appropriate where the limit is tight.
Packaging and Shipment Protection
Shipping is where the last damage happens. A board that is packed with the fingers touching another board will abrade them in transit, and the damage is attributed to the customer’s socket rather than to the packaging. Edge protectors, interleaves and a box that does not allow movement are the standard measures, and their cost is trivial compared with a field failure at the connector.
The packaging should also exclude contamination. A cardboard box with a high sulphur content can tarnish silver finishes and can leave a film on gold, and a foam that outgasses will deposit a film on the contacts over time. Where the product has a long shelf life, the packaging materials should be checked for compatibility, and the board should be bagged with a desiccant. The packing and documentation procedure is the right place for both the protection and the record of it.
FAQ
Is gold plating on the fingers always necessary? Where the connector is inserted repeatedly, yes, because the alternative finishes oxidise and their contact resistance rises. Where the connection is made once and soldered, a gold finish on the fingers buys little and adds a cost and a compatibility concern with the solder.
Can a scratched finger be repaired? Plating cannot be restored locally in a way that matches the original hardness. Where a finger is damaged, the board is usually rejected, or the entire edge is re-plated if the shop has that capability, which is a decision that should be made against the electrical requirement rather than the appearance.
Does the bevel remove the plating? It removes it from the bevel face and leaves the plating on the contact area. The requirement is that the bevel does not tear the plating at the transition or leave a burr, both of which are visible under magnification and both of which are process rather than design problems.
