Surface Finish for Lead-Free Boards and Test Contact
A surface finish has to do two jobs that are often discussed separately. It has to keep the copper solderable until the assembly is built, and it has to present a surface that a test probe can penetrate reliably. Moving to a lead-free process changes both, and the second is the one that is most often overlooked.
This article looks at the finishes used on lead-free boards, at what happens physically when a bed-of-nails probe touches a test pad, and at the design decisions that keep the test stage working. Most of what follows is settled long before anyone sets up a fixture, which is why the finish and the test access belong in the same conversation.
Why the Finish Matters More Without Lead
Tin-lead finishes were forgiving. The alloy was soft, it reflowed at a low temperature, and it tolerated the thin oxide that formed on a pad during storage. Lead-free finishes have to achieve the same solderability with a different chemistry and a much smaller process window, and some of them are far more sensitive to storage and handling than the finish they replaced.
The choice therefore propagates further than the solder joint. It affects the shelf life of the bare board, the number of reflow passes the surface can survive, the contact behaviour during test, and the appearance criteria that inspection will use. The move from leaded to lead-free assembly is a process change rather than a material substitution.
Comparing the Available Finishes
Organic solderability preservative is the cheapest option and the easiest to apply, and it protects copper with an organic layer rather than a metal one. It survives a limited number of reflow cycles and is difficult to measure, so it suits boards that will be built promptly and inspected visually rather than by probe contact.
Immersion silver and immersion tin offer good solderability and flat surfaces at moderate cost, but both are thin and both are sensitive to handling and to storage conditions. Electroless nickel with immersion gold is the finish most often chosen where the surface has to remain solderable and also present a durable contact, because the gold layer protects the nickel underneath. Hard gold plating, used on edge connectors and on pads that will be probed repeatedly, is a different process again, and the way these coatings are deposited relates directly to the defects seen in plated copper.
How an ICT Probe Makes Contact
An in-circuit test fixture brings a bed of spring-loaded probes down onto the board, and each probe has to make a low-resistance connection to a test pad. The probe tip is much harder than the finish on the pad, and the contact pressure it applies is well above the yield strength of the solder or the coating.
What happens next is mechanical. The tip deforms the surface, penetrating through whatever contamination or oxide has formed on it, until it reaches clean metal underneath. The quality of the contact therefore depends on the finish being soft enough to deform and on the tip being able to reach metal rather than sitting on an insulating layer. The contact resistance measured by the tester depends on how much of the tip is in contact with clean material.
Contact Resistance and What Changes It
A finish that forms a hard, chemically stable surface can resist penetration, and the probe ends up sliding across it or resting on it rather than cutting in. Resistance rises, and the variation between probes on the same board rises with it, which shows up as intermittent failures that appear and disappear between runs. The engineer then has to decide whether the fault is in the board or in the fixture, and the answer is rarely obvious from the reading alone.
Contamination makes the same problem worse. Flux residue that was not cleaned, handling residue from fingers, and the organic layer of an OSP finish all sit between the tip and the metal. Where a design relies on test access, the assembly process has to include a cleaning step that is matched to the finish, and the test engineer has to know which finish was used before deciding what a marginal reading means.
Test Pads: Size, Spacing and Mask
A test pad needs to be large enough that the probe lands on it despite the tolerance of the fixture and the board. The pad also has to be free of solder mask, with a defined opening, and it must not be shared with a component pad unless the process allows it, because probing a joint that a part occupies is a reliable way to damage the joint.
Spacing between pads is set by the probe pitch the fixture can achieve, and the pads should be placed where the probe can reach them without the tip touching a tall component. Keeping a clear area on the underside for the fixture is a design decision, and it belongs with the general rules that keep a board manufacturable and testable.
Testability as a Design Requirement
Test access is often treated as something added after the layout is finished, and it is then squeezed into whatever space remains. The result is pads that are too small, too close together or too near a component, and a fixture that produces readings the engineer cannot trust.
Treating test access as a constraint from the beginning costs a few square millimetres of board area and saves the argument that follows a marginal test result. It also means the finish can be chosen with the test stage in mind, which matters more in a lead-free process where the surface behaves differently from the alloy it replaced. The characteristics that distinguish a well-made board include that kind of traceable, repeatable access.
Managing the Finish Through the Build
The finish is not only a design choice; it carries storage and handling requirements with it. An immersion silver or an organic finish has a shelf life measured in months, and the clock starts when the panel leaves the plating line. Packaging, desiccant, and the interval between the bare boards arriving and the assembly date all belong in the plan rather than in a drawer.
Where the same design is built with a different finish at some point in its life, the test program may need revisiting, because readings that were accepted with one surface are not necessarily produced by another. Recording the finish on the fabrication drawing and in the assembly documentation is what makes that comparison possible, and quality characteristics that travel with the board are what make it useful.
FAQ
Does the finish affect functional performance? Not directly. What it affects is solderability, contact behaviour and the number of reflow passes the surface can take, all of which appear as yield rather than as electrical specification.
Can a test pad be covered with solder mask? No, the probe has to reach metal. A mask opening slightly larger than the pad gives the probe room without exposing neighbouring copper.
Is a gold finish always the best choice for test? Gold over nickel gives a durable contact, but it costs more than the alternatives and it is not needed where the board is probed once. The choice should follow the number of contacts the pad will see.
