Conformal Coating Masking and Keep-Out Control
Conformal coating protects a board only where it is allowed to stay, and the parts of the assembly that must remain clear of it are the ones that cause the most trouble. Masking defines the boundary, and a keep-out zone that is too small, or a mask that lifts during coating, produces a board that looks coated and fails at the connector.
Why Masking Decides the Coating Outcome
Coating is applied as a liquid that flows wherever surface energy takes it, so the boundary between coated and uncoated areas is set by the mask rather than by the applicator. If the mask edge lifts, coating wicks underneath and reaches a surface that was supposed to stay bare, and the defect appears only when the connector is mated or the test point is probed.
Masking also determines how much of the assembly has to be reworked. A coating that enters a connector body cannot be removed without damaging the plating, whereas a coating that stops a millimetre short of the connector is a cosmetic finding at worst. The keep-out geometry is therefore a reliability decision, not a finishing detail. A drawing that shows the coated area but not the keep-outs leaves the boundary to be decided at the bench, and two operators will decide it differently.
The Keep-Out List: What Must Stay Bare
The list is short but absolute: connector mating surfaces and contacts, test points and probe pads, mechanical mating surfaces and mounting holes, adjustable components such as trimmers, heat sinks that must bond to a thermal pad, and any surface where a later solder operation will occur. Each of these fails in a different way when coated.
Distances matter as much as the list. A keep-out of 1 to 3 mm around a connector is common, and the value should come from the connector drawing rather than from habit. Where the coating has to stop within a millimetre of a pin, the applicator needs a mask or a program that can hold that tolerance, which changes the process chosen. On a dense board, a keep-out that a programmed path cannot hold may force a hand-mask, which is slower and less repeatable than the selective program it replaced.
Tape and Boot Masking
Masking tape must adhere well enough to resist the coating but release cleanly afterward. Silicone adhesives are the usual choice because they leave little residue and tolerate the curing temperature, while rubber-based tapes can leave a gum that interferes with later soldering. The tape should be applied with a roller so the edge is sealed, since a lifted edge is the most common cause of coating creep.
Boots and caps cover connectors and threaded features more reliably than tape and are reusable if they are cleaned. Where a boot is used, the fit must be tight enough that coating cannot seep in but loose enough that removal does not damage the connector. Damaged or stretched boots should be replaced rather than reused, because a poor fit is invisible until the coating has cured.
Dispensing and Selective Coating Around Masks
Selective coating equipment follows a program, so the mask defines the boundary only approximately and the program defines it exactly. The nozzle path should be taught against the actual masked board rather than against the drawing, and the offsets recorded so a program can be reproduced after a changeover. Valve type and nozzle diameter set the width of the line the machine can lay down.
Where the coated area is close to a keep-out, the program usually needs a slightly conservative boundary, and the coating can be built up in two passes rather than one. Two thin passes cover better at the edge than one heavy pass, which tends to flow past the mask and produce a thick band that takes longer to cure.
Mask Removal and Residue
Removal is done after the coating has cured enough to hold its edge but before it becomes brittle, and the timing follows the coating specification rather than the shift schedule. Pulling tape from an uncured film drags coating across the keep-out area; pulling it after full cure can lift the coating edge and leave a ragged boundary.
Adhesive residue left behind must be removed with a solvent compatible with both the coating and the component, and the removed area should be inspected for coating that wicked under the mask. Residue on a connector contact is conductive enough to matter in some cases and should never be assumed harmless.
Coating Around Connectors and Tall Parts
Connectors and tall components shield the board beneath them from the spray, so the area directly behind a connector is often starved. Where coverage is required under or behind a tall part, the coating has to be applied from more than one direction, or the part must be masked and the area coated by hand. Coverage figures that assume a flat board do not apply here.
Coatings also wick into connector bodies by capillary action, which is a function of the gap between the contacts and the housing rather than of the application method. Where a specification calls for a connector to remain clear of coating, the correct control is a mask or a boot, not a change in spraying distance.
Cure Verification
Cure is a process condition that has to be measured, not inferred from the clock. A coating that is tack-free may still be uncured inside, and an undercured film provides little protection while looking perfect. Common checks are a solvent rub, a hardness test, or an infrared or thermal measurement against the coating specification.
The cure schedule depends on the chemistry: acrylics cure quickly, urethanes need longer and often need moisture, and UV-cured materials need line of sight to every coated surface. Where a shadowed area exists under a component, a UV-only cure leaves it uncured, which is why coatings formulated for UV cure usually include a secondary moisture or thermal mechanism.
Inspection of Coated Areas
Inspection looks for coverage, for the boundary against each keep-out, and for defects such as bubbles, voids and dewetting. A UV tracer in the coating makes the film visible under a black light, which turns a subjective judgement about coverage into an image that can be compared between boards and between operators.
Thickness is checked on a sample, either by a non-destructive gauge or by a section, and recorded against the specification. Coverage percentage and thickness are separate requirements: a film that covers everything at half the required thickness protects less than a complete film at full thickness, and both need to be verified.
Records and Rework
Record the coating material, batch, cure conditions, masking method and inspection result for each lot, because a coating defect may not appear until the board is in the field. Where rework is needed, the coating must be removed locally, the repair coated again, and the cure repeated, and that repair belongs in the record rather than being treated as a touch-up.
Rework limits deserve a number as well. Repeated coating and stripping cycles damage the mask and the laminate, so a board that has been reworked several times should be reviewed before it is coated again, and the limit written into the process rather than left to judgement.
FAQ
How large should a keep-out zone be around a connector? Usually 1 to 3 mm, taken from the connector manufacturer’s recommendation rather than habit. Where coating must stop closer, the applicator needs a mask or a taught path.
Why does coating creep past masking tape? The edge is not sealed. Applying tape with a roller and replacing stretched boots prevents the wicking that puts coating on contacts and test points.
How is conformal coating cure verified? By measurement rather than by clock: a solvent rub, hardness test or infrared reading against the coating specification, because a tack-free film can still be uncured inside.
