Conformal Coating Rework and Repair Practice

A conformal coating protects a board by sealing it, which is exactly what makes repair difficult. Once a unit is coated, any component replacement requires the coating to be removed locally, the repair to be made and the coating to be restored so that the protection is continuous. Done badly, the repair becomes the weak point that fails in the field. This article covers the removal methods, the masking and reapplying practice and the checks that confirm the repair is sound.

Why Coating Repair Is Different

The coating is part of the product, not a finish on top of it. Cutting into it breaks the barrier, and a repair that leaves an edge of old coating next to new material creates a boundary where moisture can collect. The interface between the two layers is the part of the repair that decides whether it holds.

The second difference is thermal. A repair that involves soldering heats the surrounding coating, and some chemistries soften, blister or discolour when heated. The removal area has to be large enough that the soldering operation does not affect the coating that is meant to stay.

Coating Removal Methods

Whichever method is used, the objective of coating removal is the same: expose the repair area completely while leaving the surrounding protection intact and undamaged. The area removed should be larger than the component being replaced, because the solder operation needs clearance and because a coating edge that is too close to a joint will be heated and may lift during the repair.

Chemical removal uses a solvent or a stripping gel that softens the coating so it can be lifted. It is gentle on the board and the components, and the risk is that the chemistry attacks the mask, the labels or a plastic component body. The solvent has to be matched to the coating chemistry, and the dwell time matters as much as the product.

Mechanical removal uses abrasion, scraping or a fine tool to cut the coating away. It is faster and more localised, and the risk is damage to the mask, a lifted pad or a scratched trace. Localised thermal removal is a third option, using a controlled heat source to soften the material before it is lifted, and it is useful where the coating is thick but requires care around sensitive parts. The protection and application background of the material is described in conformal coating in PCB assembly.

Masking for a Local Repair

Masking defines the boundary of the repair. Everything that must remain coated is protected while the work is done, and everything inside the repair area is exposed so that the soldering operation does not trap material under a component. The mask should be placed so that the new coating overlaps the old by a defined margin, because a butt joint between two layers is weaker than an overlap.

The overlap margin should be specified rather than improvised. A few millimetres of overlap gives the new material something to bond to and covers the damaged edge of the old coating, while a repair that stops exactly at the old edge leaves a seam. Masking material itself must not leave residue, since a silicone or an adhesive residue on the surface will prevent the new coating from adhering.

Soldering Inside a Coated Board

Soldering on a board that carries coating has to be done with the surrounding area protected. Flux and solder residue must be removed before the new coating is applied, because any contamination under the new layer becomes a conductive path along the interface and a corrosion site. Cleaning is therefore a required step and not an optional one, and the cleaning chemistry must be compatible with the coating that remains.

Thermal exposure should also be limited. A hot air tool that heats the surrounding coating can blister it, and the damage may only appear after the board returns to service. Directing the heat to the joint and keeping the removal area large enough that the boundary is not heated is the practical rule.

Reapplication and Cure

Reapplication has to match the original. The same chemistry, the same thickness and the same cure are needed if the repaired area is to behave like the rest of the board, and a different material applied over the old coating may not bond. Where a substitute has to be used, it should be qualified on a sample before the repair is attempted on a real unit.

Cure is the step most often cut short. A repair area that cures at room temperature takes hours to reach full properties, and one that is handled before it cures can be disturbed or contaminated. UV curing is faster but only reaches the areas the light can see, so a shadowed region under a component may remain uncured unless a secondary cure mechanism is present.

Adhesion Between Layers

Adhesion between the old coating and the new is the property that decides whether the repair holds. It is affected by the surface condition of the old coating, by the overlap geometry and by the cure of the new material. A surface that has been contaminated by flux, by masking residue or by handling will not bond, and the failure appears later as a lifted edge.

Testing should be local and non destructive where possible. A tape test on a sample of the same repair, or a visual check of the film for lifting at the boundary, gives an indication. Where the product requires it, a test coupon coated and repaired with the same process can be used for a more formal assessment. The general inspection logic used for coating is described in conformal coating inspection.

Documentation and Process Control

Every coating repair should be recorded with the removal method, the area, the materials used and the cure. That record is what allows a repaired unit to be treated as a known quantity rather than as an unknown risk, and it is what makes it possible to spot a repair process that is producing repeat failures.

The gopcb assembly group treats coating repair as a defined process with its own procedure, tooling and training. Where a repair cannot be made without damaging the surrounding protection, the honest answer is that the unit should be reworked as an uncoated assembly rather than patched. Comparing the repaired result against the criteria used for new coating, described in judging PCB quality, keeps the standard consistent.

Restoring Protection Around Connectors and Openings

Repairs near a connector or an opening need extra care, because the coating in those areas is often applied with a different requirement. A connector that must remain free of coating, a shield that needs a conductive path or an opening that must not be sealed all define regions where the repair has to stop short. Those boundaries should be identified on the drawing so that the technician does not have to interpret the requirement at the bench.

Where a repair crosses such a boundary, the work should be split into the part that can be coated and the part that must not be, with a mask that is removed only after the coating has cured. Handling a masked area before cure, or removing the mask too early, produces a ragged edge that is both cosmetic and functional, because a ragged edge is where a coating lift usually begins.

FAQ

Can a different coating be used for the repair? Only if it is qualified to bond to the original chemistry. Mixing chemistries is a common cause of a repair that lifts within weeks.

How much overlap is needed between new and old coating? A few millimetres of clean overlap is the usual practice. The exact figure should come from the coating supplier and from a test repair.

Does a repaired area need to be inspected differently? Yes. The boundary between the layers is the critical region, so inspection should focus on the overlap rather than on the centre of the repair.