Solder Mask Adhesion, Wicking and Cure Control
Solder mask is a permanent coating rather than a paint, and it has to do three jobs at once: protect the copper, define the areas that will be soldered, and survive assembly. Adhesion and wicking are the two ways it fails, and both start in the preparation rather than in the coating.
Adhesion Starts at the Copper
The mask bonds to the copper and to the laminate, and the two surfaces are prepared differently. Copper that has been oxidised or that carries an organic residue gives a bond that fails at the interface, and laminate that has been over-etched or left with smear gives a similar result.
Preparation is therefore the first variable rather than the last. A pumice or a chemical treatment is used to give a controlled roughness, and the roughness has to be consistent across the panel for the adhesion to be consistent. Our board quality notes describe how that is judged.
Wicking and What Causes It
Wicking is mask that has crept into a hole or under a lifted edge, and it appears where the mask has been pushed rather than where it was printed. It usually follows an incomplete cure, a screen or a spray that placed too much material, or a hole that was not cleaned before coating.
Once wicking has occurred it cannot be corrected by curing harder, because the material has already moved. The correction is on the coating side, and it is found by looking at where the wicking appears rather than at how much of it there is.

Tenting and Via Coverage
Tenting is mask over a via, and it is a deliberate choice on many designs. A tented via is protected and a via that is left open can be probed, but the two decisions lead to different process settings and different inspection.
A tent that fails becomes a lifted mask over a hole, which is a defect that appears after reflow rather than before. Where vias are tented, the adhesion over the hole is the item to verify. Our warpage notes cover the related behaviour of the panel during cure.

Cure and Its Measurement
Cure is measured rather than assumed, because a mask that is under-cured will fail in the assembly oven and an over-cured one becomes brittle and chips at the edges. The usual measurement is a solvent rub, a pencil hardness test or a spectroscopic method, and any of the three is better than a temperature log alone.
The cure also has to survive the assembly profile, which is hotter than the cure. Where the two are close, the mask will continue to cure in the reflow oven and its adhesion over the thermal excursion is the item that decides whether it survives. Our coating inspection notes describe the same idea for a coating applied later.
Registration and the Solder Dam
The mask is registered to the pads, and the registration decides the solder dam between them. A dam that is too thin will not hold the paste during reflow, and a dam that is too wide takes area the pad could have used.
Registration error is the item that reduces a comfortable dam to a marginal one, so the mask exposure and the pad imaging have to be held together. The tolerance belongs on the drawing as a stated dam width rather than as an assumption.
Finishes and Their Interaction
The mask is applied before the surface finish in most flows, so the finish sees the mask edge. A finish that creeps under the mask lifts it, and a mask that is not fully cured can be attacked by the chemistry, so the two processes are developed together rather than in sequence.
Where the finish is a hot one, such as a silver immersion or a thick gold, the mask edge is exposed to the chemistry for longer and the adhesion requirement is higher. That is a process decision rather than a design one, and it is worth stating on the fabrication notes.
Inspection and Acceptance
Inspection covers the dam width on a coupon or on the panel, the adhesion by tape test, the cure by one of the chemical methods and the appearance of the edges. Cross sections are used where the mask has to be judged at a step or over a via.
Keeping a sample from the panel turns a later dispute into a measurement. The sample is worth having even where the acceptance test is simple, because it lets the same part be re-examined without rebuilding the batch. Our joint criteria notes describe how the assembled result is judged.
Process Control and Verification
On a design of this kind, wicking is the item that decides how the rest of the board is arranged. Keeping a sample from the panel turns a dispute into a measurement, because the same coupon can be re-examined by both parties without rebuilding the batch. Reviewing the design before the data is released is cheaper than correcting it after the panel is in the tank, because every step downstream inherits the decision made at the front end.
Documenting the assumption is part of the design work, and a short note on the drawing prevents a question that would otherwise arrive a day later and cost a day of schedule. The process window is set by the narrowest step in the flow, so an improvement anywhere else shows up as margin rather than as yield until that step is addressed.
A short note on the drawing about handling, storage or packaging is often worth more than an extra decimal place on a tolerance.
Process Control and Verification
On a design of this kind, wicking is the item that decides how the rest of the board is arranged. Keeping a sample from the panel turns a dispute into a measurement, because the same coupon can be re-examined by both parties without rebuilding the batch. Reviewing the design before the data is released is cheaper than correcting it after the panel is in the tank, because every step downstream inherits the decision made at the front end.
Documenting the assumption is part of the design work, and a short note on the drawing prevents a question that would otherwise arrive a day later and cost a day of schedule.
Process Control and Verification
On a design of this kind, wicking is the item that decides how the rest of the board is arranged. Keeping a sample from the panel turns a dispute into a measurement, because the same coupon can be re-examined by both parties without rebuilding the batch. Reviewing the design before the data is released is cheaper than correcting it after the panel is in the tank, because every step downstream inherits the decision made at the front end.
FAQ
Does a thicker mask protect better? Up to a point. A thick mask is harder to cure fully and it flows more at the dam, so it is not automatically an improvement.
Can wicking be removed after cure? It can be scraped in some places, and doing so leaves a damaged edge that will fail later. Preventing it is the practical answer.
What does gopcb provide for solder mask control? We provide surface preparation set for the copper and the laminate, coating and cure parameters recorded against the adhesion test result, dam width verification against the drawing, coupon retention, and inspection records for each batch.



