PCB Solder Mask: Purpose and Design Rules

Solder mask is the coloured coating on a finished board, and its appearance is the least important thing about it. The mask controls where solder can go, protects the copper from the environment and defines a set of spacing rules that a design must respect. Getting those rules wrong produces defects that appear during assembly and are attributed to something else.

What the Mask Does

The primary function is to prevent solder from bridging between adjacent joints. During reflow the alloy is liquid and free to move, and without a mask between two pads a small excess of paste can join them permanently.

The second function is protection. Copper oxidises and corrodes, and a mask that covers the traces removes them from contact with the environment. That protection matters most where the board will see humidity, contamination or condensation, and it is part of the reason the mask must adhere well rather than merely cover. Our component tolerance and reliability notes describe how the coating affects long term behaviour.

Mask Dams and Their Limits

The web of mask between two adjacent pads is called a dam. Although mask is an insulator, a dam formed between fine pitch pads is thin and can be lifted or washed away during assembly, which restores the bridging risk it was meant to prevent.

Fabricators therefore specify a minimum dam width, and it is typically much larger than the minimum copper spacing. A layout that meets the copper spacing rule can still violate the mask rule. Checking pad spacing against the mask dam requirement rather than against the copper rule is one of the differences between a design that assembles cleanly and one that does not.

Registration Tolerance

The mask is imaged and developed like any other layer, so its position has a tolerance. The openings are therefore drawn larger than the pads they expose, so that a small misalignment still leaves the whole pad uncovered.

That enlargement is the reason mask openings are not simply pad shaped and sized. It also means the exposed area is larger than the pad, which affects the paste deposit on features where the paste is printed to the pad size. Our stencil fabrication notes describe how the paste layer accounts for it.

Vias and Tenting

A via that is not used as a test point or a solder feature can be covered by the mask, a practice known as tenting. Tenting protects the via from contamination and prevents solder from being drawn into the barrel during assembly.

Full tenting, in which the mask covers both ends, is not always achievable because the mask can slump into a large via opening. Where the via is large or where the assembly process is aggressive, the via is plugged with a filling material before the mask is applied, or it is left open and treated as a feature. Our via placement article describes the cases where each approach applies.

Interaction With Surface Finish

The mask is applied and cured before the surface finish on most processes, which means the finish bath sees the mask. A mask that is not fully cured can be attacked by the chemistry, and the resulting loss of adhesion appears as lifting after assembly.

The finish also determines how the exposed copper behaves. A mask that creeps over the edge of a pad during development leaves part of the copper unprotected, and the finish may not adhere properly to that sliver. Where a pad is marginally small, the missing sliver can be enough to impair the joint. Our ENIG surface finish article describes one finish where the interaction is particularly important.

Adhesion and Its Causes of Failure

Mask adhesion depends on the surface preparation before application. Contamination, residual oxide or an incompletely cleaned copper surface all reduce the bond, and the failure appears later rather than immediately.

Thermal cycling is what usually reveals it. A board that has been through assembly without incident may show mask lifting after a few thermal cycles, because the mask and the laminate expand at different rates and the bond is what holds them together. Where a mask lifts, the copper beneath is exposed and the protection the mask was supposed to provide is gone.

Colour and Its Practical Effects

Colour is a cosmetic choice with two practical consequences. A glossy surface reflects light into the inspection camera, which can obscure a joint or create a false indication, while a matte surface scatters it and gives a more even image.

The colour also affects the contrast between the mask, the copper and the silkscreen. A white silkscreen on a black mask provides high contrast for operators and for automated vision, while a dark legend on a dark mask does the opposite. The choice should be made with the inspection method in mind rather than for appearance alone.

Design Rules Worth Checking

Before a design is released, the mask layer should be reviewed against the fabricator capability. The following checks catch the majority of the problems that would otherwise appear during assembly:

• Confirm the minimum mask dam width against the fabricator figure, not against the copper spacing rule.
• Check every pad for a mask opening; a pad without one cannot be soldered.
• Check for mask openings over features that should not be exposed, such as traces and plane areas.
• Confirm that mask openings extend beyond the pad on every side by the registration allowance.
• Verify that vias to be tented are small enough for the mask to bridge reliably.
• Confirm which vias are plugged, which are tented and which are left open, and that the assembly drawing says so.
• Check that the silkscreen does not overlap any mask opening or pad.
• Confirm the mask colour choice against the inspection method used in production.
• Check that no mask opening encroaches on the board edge.
• Verify that the mask layer is complete and that no copper feature was missed during the export.

Each of these takes minutes to check and each of them has been the cause of a production problem on some board. Our design release checklist places them in the review sequence.

Mask Defects and What They Look Like

Most mask problems are visible before assembly if the board is examined against a light, and all of them are cheaper to catch at that point. The common defects and their appearance are worth knowing when incoming material is checked.

A dam that is missing or broken appears as a narrow gap where two pads are connected by exposed copper. Lifting appears as a lifted edge with a shadow beneath it, usually starting at a via or at the edge of a large copper area. Residue left in an opening appears as a dull film over the pad, which will prevent the paste from wetting. Our surface finish notes describe how that film interacts with the finish.

FAQ

Is the minimum mask dam the same as the minimum copper spacing? No, and it is usually larger. The two rules come from different processes and both must be satisfied.

Should all vias be tented? Vias that are not used for test or soldering are normally covered. Large vias may need to be plugged instead, because the mask cannot bridge the opening reliably.

Why does mask lift after assembly? Usually because the surface preparation before application was inadequate, or because the mask was not fully cured. The failure appears after thermal cycling rather than immediately.

Does mask colour affect soldering? Not directly. It affects inspection, because it changes how the surface reflects light.

Can mask replace a conformal coating? No. Mask covers the copper; it does not seal the assembly against moisture and contamination.