Overmolding PCB Design: Encapsulation and Material Choice

Overmolding turns a circuit board into a sealed component. Instead of a housing with a board inside it, the board is placed in a mold and plastic or a soft encapsulant is formed around it, so the finished part is a single sealed object with no gasket, no screws and no air gap. The technique is used for sensors, connectors, motor controls and anything that has to survive water, dust or a washdown.

Why a Board Is Overmolded

The mechanical reasons are the strongest. An overmolded part has fewer joints to leak, it resists impact and vibration better than an assembly of parts, and it removes the fasteners and seals that would otherwise be the first items to fail. The thermal reason is also real: the encapsulant conducts heat away from the components more effectively than air does, which allows a higher power density in the same volume.

The economic reason is assembly cost. Molding replaces a housing, a gasket, several fasteners and the labour to fit them with a single process step. That is why the technique dominates in high volume products where the tooling cost is recovered many times over.

Low Pressure Molding and Rigid Encapsulation

Low pressure molding uses a hot melt polyamide or polyolefin injected at a modest pressure and temperature. The low pressure means the material flows around delicate components and wire bonds without displacing them, and the low temperature means the board and the solder joints are not stressed during the process. It is the technique of choice for small sensors, cables and connectors.

Rigid encapsulation uses an epoxy or polyurethane that is poured or injected and then cured. It gives a harder, more chemically resistant shell and better thermal performance, at the cost of higher stress on the components during cure and, in most cases, no rework path. The choice between them follows from the mechanical environment and from whether the part is expected to be repairable.

What the Process Does to the Board

Molding is not a neutral operation for the circuit. The material shrinks as it cools or cures, and the shrinkage applies a compressive load to the board and to everything on it. Laminate and copper expand and contract at different rates, and a board that is unbalanced in its layer construction will bow under that load. Components see the same stress, and a large ceramic capacitor placed near the edge of the panel is a common casualty.

The thermal load matters as well. Even a low pressure process raises the board to a temperature where the encapsulant flows, and any moisture absorbed in the laminate will expand and cause delamination or blisters. Boards destined for molding are therefore baked before the process, and the stackup is kept symmetric, in line with the material behaviour described in PCB dimensional stability and expansion.

Design Rules for a Molded Board

Design for the flow first. The encapsulant needs a path around the board, and tall components should be placed so that they do not create a void behind them where the material cannot reach. Voids are the most common molding defect and the hardest to detect, because they are invisible from the outside and only appear later as a leak path or as a thermal hot spot.

Allow for the shrinkage and for the tool. The board outline should have enough tolerance for the mold cavity, the features that locate the board in the tool should be placed where they will not interfere with the circuit, and any connector or cable exit should be designed so that the material seals around it rather than relying on a mechanical clamp. The general principles of outline and mounting features are described in board outline and mounting design.

Materials and Adhesion

The encapsulant has to bond to the board, to the components and to any insert. Solder mask chemistry, surface contamination and the finish on the board all affect that bond, and a release agent left on a surface is enough to produce a delamination that shows up in a thermal cycle. The material and the board surface should be qualified together rather than separately.

Adhesion promoters and plasma treatment are used where the bond is critical. The choice of encapsulant also determines the temperature range of the finished part and its resistance to chemicals, so the material is selected against the environment rather than against the immediate process. Where a full encapsulation is not required, a conformal coating may be sufficient and is far easier to apply and to rework, as described in conformal coating for board protection.

Thermal Path and Component Placement

An encapsulant conducts heat far better than air, so a molded board can run hotter components than the same circuit in a ventilated housing. The heat leaves the component through the solder joint into the copper, spreads through the board and then passes into the molding compound and out to the surface. Each step in that chain can be the bottleneck, and the one that is usually overlooked is the copper area under the component.

Placement follows from the same reasoning. Power devices should sit where the copper is heaviest and where the molded material has a short path to the outside of the part. Keep temperature sensitive components such as crystals, electrolytic capacitors and sensors away from heat sources, and remember that after molding there is no convection to help, only conduction. A thermal simulation before the mold is cut is far cheaper than discovering a hot spot afterwards.

Sealing Connectors and Cable Exits

The board is only one part of the seal. A cable leaves the molded body through an opening, and that opening has to be closed by the molding process itself rather than by a separate grommet. The cable jacket and the molding compound must bond, which means the jacket material has to be compatible with the encapsulant and clean when the part is molded.

Connectors present the same problem in a different form. A connector that is overmolded has to survive the temperature and pressure of the process, and its contacts must not be displaced by the flow. Where a connector cannot tolerate the process, the alternative is to mold around a sealed feedthrough or to leave a cavity that is closed later, both of which add a step and a potential leak path.

Testing and Failure Modes

The failures that matter after molding are leaks, voids, delamination and cracked components. A leak is detected by pressure or by immersion testing, a void is found by X-ray or by acoustic imaging, and delamination is found by thermal cycling followed by a section or an acoustic scan. Cracked components are usually found by functional test after the thermal cycle that caused them.

Reliability testing is therefore built around thermal cycling and humidity, because the molded part is normally exposed to both and because the interface between the encapsulant and the board is where the failures concentrate. The test programme should reproduce the actual temperature range and the actual number of cycles the product will see, and it should be run on parts molded with the production tool rather than on hand poured samples.

FAQ

Can a molded board be repaired? Usually not without destroying the encapsulation. Low pressure molding can sometimes be peeled away, but a rigidly potted assembly is normally scrapped.

Does overmolding replace a conformal coating? It replaces it when full sealing is needed. A coating protects the surface but leaves air around the components and does not seal a connector.

Why are boards baked before molding? To remove absorbed moisture, which would otherwise expand during the process and cause blisters or delamination in the laminate.