Between two adjacent surface-mount pads, a thin wall of solder mask, the dam, keeps molten solder from flowing across the gap and shorting the pads. When the dam is too thin or missing, solder bridges form during reflow, and on fine-pitch parts those bridges are the dominant assembly defect. Dam thickness is a layout decision made long before the board is manufactured, and it interacts with the fab's capabilities: a dam your layout demands but the fab cannot reliably produce is as bad as no dam at all.
How the Dam Prevents Bridging
During reflow, solder paste melts into a liquid with high surface tension that wets to exposed copper and avoids the solder mask. The dam works by presenting a mask-covered barrier between the pads: the molten solder on each pad is held back by the mask's refusal to wet, and surface tension keeps each blob coherent on its own pad. Without the dam, the two liquid solder volumes can touch across the narrow gap, and once they touch, surface tension merges them into a single bridge rather than pulling them apart. The dam does not need to be tall; solder mask is only tens of microns thick. It needs to be continuous and wide enough that the fab can actually image it: a dam drawn in CAD but too narrow to survive the fab's process is a fiction, and the bridging it was meant to prevent happens anyway.
Minimum Dam Widths and Fab Capabilities
The critical number is the minimum solder mask dam the fabricator can reliably produce, which depends on their imaging and registration tolerances. A common capability threshold is around 3 to 4 mils (roughly 75 to 100 microns): dams narrower than this may break, pinhole, or misregister, leaving gaps where bridging occurs. This threshold must drive the layout: the pad-to-pad gap minus twice the mask expansion (the clearance the fab adds around pads) must leave at least the minimum dam width. On fine-pitch parts, this arithmetic gets tight fast. A part with pads on a very fine pitch may simply not leave room for a reliable dam at a standard fab, which is the signal to either choose a more capable fab, accept a higher bridging risk with tighter process control, or reconsider the part. Designing dams below the fab's capability and hoping is the most expensive form of optimism in PCB assembly.
Layout Techniques for Reliable Dams
Several layout choices protect dam integrity. Keep solder mask expansion modest: excessive expansion eats into the gap between pads and thins the dam, while too little expansion risks mask overlapping the pad and reducing the solderable area. Follow the fab's recommended expansion values rather than the CAD default. Avoid drawing custom mask shapes that create slivers: let the CAD tool generate mask from the pads with uniform expansion, which produces the most manufacturable result. In dense areas, consider slightly reducing pad widths within the part's tolerance to buy gap width for the dam; a marginally narrower pad solders fine, while a missing dam bridges reliably. For the finest pitches, some designers use mask-defined pads, where the mask opening is smaller than the copper pad, which guarantees dam material between pads at the cost of tighter registration demands.
When Dams Cannot Be Had
Some geometries defeat dams entirely: ultra-fine-pitch parts, certain connector footprints, and dense BGA fanouts where the gaps are simply too small for any producible dam. In these cases, bridging prevention shifts to process control. Reduce the solder paste volume with smaller stencil apertures, since less solder means less material available to bridge. Use a paste with appropriate particle size for the pitch; coarse paste on fine pitch is a bridging guarantee. Optimize the reflow profile: a well-controlled soak and ramp gives flux time to clean and lets surface tension do its work, while a rushed profile leaves bridges. Nitrogen reflow improves wetting and reduces bridging on difficult geometries. And plan for inspection and rework: fine-pitch assembly without dams should assume some bridges will occur and budget the rework time, rather than discovering it on the production line.
Verifying in Gerbers
Before sending the board out, inspect the solder mask Gerber layer directly. Zoom into the finest-pitch areas and look at the dams: they should be continuous, reasonably uniform strips of mask between every adjacent pad. Look for slivers, pinholes, and places where the mask openings of neighboring pads have merged into one opening, which means no dam at all. Check the fab's DFM report if they provide one; many fabs flag dam violations automatically. This inspection takes minutes and catches the layout errors that cause the most rework. A board with verified dams assembles cleanly; a board with assumed dams assembles expensively.
The solder mask dam is a tiny wall doing a critical job. Size it for what the fab can actually build, protect it with sane mask expansion, verify it in the Gerbers, and have a process plan for the geometries where dams cannot exist. Bridging is a layout problem first and a process problem second; solve it in layout.
Related Guides
- Copper Land Pad Widths for Hand Solder Clearances: Hand soldering is more forgiving than reflow in some ways and less in others: the iron can reach places the oven cannot, but the human hand needs
- Selective Solder Drag Speeds on Dual-In-Line Leads: Drag soldering a row of dual-in-line leads looks like a magic trick the first time you see it done well. A grooved tip carrying a small reservoir of
- Selective Solder Flux Wave Heights on Mixed Boards: Selective soldering solders through-hole components on a board that already carries surface-mount parts, using a small, precise solder wave instead
- Thermal Relief Spoke Sizing on Power Planes: Thermal reliefs are the compromise between two enemies: the soldering iron, which needs the joint hot, and the power plane, which wants to wick that