PCBA Shielding Material Processing: Usage Specs That Actually Contain EMI

Shielding on a PCBA is not about wrapping the board in aluminum foil and hoping for the best. It is a precision engineering task. If the shielding material does not make proper electrical contact with the ground plane, if the gasket is compressed unevenly, or if the conductive coating has a resistance higher than specified, you have not shielded anything. You have just added weight and cost to a board that still fails EMC compliance.

Electromagnetic interference does not care about your enclosure. It leaks through gaps, radiates from cable exits, and couples through slots in the shielding can. The shielding material is only as good as how it is processed and installed.

Conductive Gasket and Finger Stock Selection

Contact Resistance Determines Shielding Effectiveness

A conductive gasket between the PCB shield can and the main enclosure looks simple. It is just a strip of metal with a foam core. But the contact resistance at the mating surface is what actually blocks the RF energy. If the resistance exceeds 2.5 milliohms across the joint, the shielding effectiveness drops by 20 dB or more at frequencies above 1 GHz.

Specify gaskets with a contact resistance below 1 milliohm at the actual compression force you will apply. The compression force matters because the gasket must deform enough to fill surface irregularities on both the can and the enclosure flange. Too little compression leaves air gaps. Too much compression crushes the foam core permanently, and the gasket never recovers its contact force after thermal cycling.

Verify the compression range on every incoming lot. A gasket that is 5% out of spec on thickness will either not compress enough or will bottom out, and neither condition gives you shielding.

Finger Stock Height and Placement Tolerance

Finger stock (also called EMI gasket fingers or beryllium copper fingers) provides the electrical bridge between the shielded can and the PCB ground plane. The finger must compress against the PCB copper when the lid is closed. If the finger is too short, it does not touch. If it is too tall, it pokes through the solder mask and shorts signal traces.

The height tolerance for finger stock is typically ±0.1mm. Place fingers within 2mm of every seam and slot in the can. The spacing between fingers should not exceed 15mm along any edge. For high-frequency applications above 3 GHz, reduce the spacing to 8mm to prevent slot radiation.

Do not place finger stock near high-current traces or components that generate heat. The finger can act as a heat sink for the component, changing its thermal behavior, or it can create a creepage path that violates safety clearance requirements.

Conductive Coating and Plating Specifications

Spray Coating Thickness and Uniformity

Conductive spray coating (nickel, copper, or silver-loaded paint) is often used to shield internal cavities or to cover non-conductive enclosures. The coating thickness directly affects shielding effectiveness. A thickness of 25 to 50 micrometers gives about 30 to 40 dB of shielding at 1 GHz. Below 20 micrometers, the effectiveness drops sharply.

Uniformity is the real challenge. A coating that is 50 micrometers on one side and 10 micrometers on the other gives you the shielding of the thin side. Use a calibrated thickness gauge to measure coating on every panel. The variation across a single board should not exceed ±10 micrometers.

The surface resistance of the cured coating must be below 0.1 ohms per square for nickel-based systems and below 0.01 ohms per square for silver-based systems. Test this with a four-point probe after every batch cure. A coating that looks shiny but measures 0.5 ohms per square is decorative, not functional.

Plating on Plastic Enclosures

Electroless nickel plating on plastic housings is common for mid-range shielding. The nickel layer must be at least 3 to 5 micrometers thick to be continuous and pinhole-free. Thinner plating has micro-pores that let RF energy leak through.

The adhesion of the plating to the plastic substrate is critical. A cross-hatch adhesion test should show no delamination. If the plating peels during assembly, the shielding is gone. Specify a minimum adhesion rating of 4B on the cross-hatch test.

For plastic enclosures that require higher shielding, consider a two-layer approach: electroless nickel for adhesion and corrosion resistance, followed by a thin copper or silver overplate for conductivity. The overplate adds less than 2 micrometers but can improve shielding effectiveness by 10 to 15 dB.

PCB-Level Shielding: Cans and Clips

Shield Can Solder Joint Integrity

A shield can soldered directly to the PCB ground plane creates a Faraday cage around sensitive circuits. But the solder joint around the perimeter is the weak link. If the joint has voids, the can is not grounded, and it acts as an antenna instead of a shield.

The solder fillet around the can must be continuous with no gaps exceeding 1mm. The joint height should be at least 0.5mm to ensure mechanical strength. Use X-ray inspection on every can to verify voiding is below 15% of the joint area. A can with 25% voiding along one edge will leak RF energy from that exact edge.

For cans with thermal pads, the solder joint under the pad must be void-free. The thermal pad serves double duty: it grounds the can and it conducts heat from the components inside. A void under the pad means poor grounding and poor heat dissipation. Both are failures.

Clip-On Shields: Mechanical Pressure Matters

Clip-on shields do not use solder. They rely on spring pressure to maintain contact with the PCB ground plane. The clip force must be sufficient to overcome the surface roughness of the PCB copper and any solder mask over the ground pads.

Specify a minimum contact force of 1.5 Newtons per clip. Test this with a force gauge during incoming inspection. A clip that has been bent and re-bent loses its spring tension. Reject any clip that does not meet the force spec.

The ground pads under the clip must be exposed copper — no solder mask. If the mask is present, the clip sits on an insulator and the shield is useless. Use a solder mask defined (SMD) pad or a mask opening that exposes at least 90% of the pad area.

Cable and Connector Shielding Integration

Shielded Cable Termination

The cable exit is the number one leakage point on any shielded enclosure. A shielded cable that enters through a hole in the metal can creates a slot antenna if the shield is not bonded to the can wall.

Use a cable gland with a 360-degree bond to the can. The gland body must make metal-to-metal contact with the can wall. The cable shield braid must be clamped under the gland with at least 10mm of braid coverage. Do not rely on the drain wire alone — it has too much inductance at high frequencies to be effective above 100 MHz.

For connectors mounted on the shielded can, use connector backshells that bolt to the can wall. The backshell provides the RF bond between the connector shell and the enclosure. A connector without a backshell is an open door for EMI.

Filter Integration at the Boundary

Shielding the enclosure is not enough if unfiltered signals cross the boundary. Every signal line that enters or leaves the shielded zone must pass through a filter — typically a feedthrough capacitor or a pi-filter array.

The filter must be mounted directly on the can wall, with one side inside the shield and one side outside. The ground terminals of the filter must bond to the can with the lowest possible inductance. Use multiple vias or a solder pad that surrounds the filter ground pins.

A filter mounted 10mm away from the can wall has 10mm of trace inductance that defeats half the filtering. Mount it flush. Solder it down. Bond it.

Processing and Handling Rules

Cleaning Before Shielding

Any contamination on the mating surface kills shielding effectiveness. Fingerprints, flux residue, dust, and oxidation all increase contact resistance. Clean the can flange and the enclosure mating surface with isopropyl alcohol before closing the shield.

For gaskets, do not touch the contact surface with bare fingers. The oil from skin creates an insulating film that takes hours to evaporate. Use lint-free gloves or handle by the edges only.

Torque Specifications for Shield Fasteners

The screws that hold the shield can to the main enclosure must be torqued to spec. Under-torquing leaves gaps. Over-torquing warps the can and breaks the gasket.

Use a torque driver set to the manufacturer's specification — typically 0.3 to 0.5 Nm for small can screws. Tighten in a star pattern, not sequentially. Sequential tightening warps the can and creates uneven gasket compression.

Re-torque after the first thermal cycle. The gasket settles and the screws relax. A second torque pass after 24 hours at operating temperature ensures consistent contact pressure for the life of the product.

Verification Testing

Do not assume the shielding works. Test it. Use a near-field probe to scan the assembled board at operating frequency. Any hotspot on the can surface indicates a leak — a bad gasket, a missing finger, or a cracked solder joint.

For full system verification, run a radiated emissions test in an anechoic chamber. The shielding should provide at least 20 dB of attenuation compared to the unshielded baseline. If it provides less, find the leak with the near-field probe and fix it before shipping another unit.