How Do You Use RF Shielding Cans and EMI Gaskets in PCB Design?

Detailed Explanation

RF shielding cans and EMI gaskets are physical shielding techniques used on real products — cellular and Wi-Fi modules routinely ship under a can, and many enclosures rely on gaskets across seams — but neither is explained anywhere on this site as its own topic; existing EMC and RF pages cover layout-based emission control in depth but not the added-hardware shielding technique that's used when layout alone isn't enough. This page covers when a shielding can or gasket is actually the right tool, and the design details that determine whether it works.

How a Shielding Can Works

A shielding can forms a conductive enclosure — a Faraday cage — around a section of the PCB, reflecting and absorbing incident electromagnetic energy so it can't couple in or out of the shielded region. Two-part cans (a soldered frame plus a removable lid) allow rework access after assembly; one-piece stamped cans are lower cost but require desoldering to access the components underneath. Both types depend entirely on the same three things to actually perform as shields:

1. A continuous, low-impedance ground connection around the entire perimeter. The can's frame must solder to a ground-connected pad pattern on the PCB with no gaps — any break in that perimeter connection is a slot antenna that leaks at the frequencies whose wavelength is comparable to the gap.
2. Via stitching beneath the ground pad pattern, connecting the perimeter ground pads down to the board's internal ground plane(s) at a spacing small relative to the wavelength of concern — without this, the ground pattern itself has enough impedance at RF to partially defeat the shield even with a perfect solder joint.
3. Minimal, deliberately sized apertures. Every opening in the can — ventilation holes, cable exits, visible-light windows — is a potential leakage path, and its effectiveness as a leak is set by its size relative to the wavelength at the frequency of concern, not by its area alone. A long, narrow slot leaks more at a given frequency than a round hole of the same area, because the slot's longest dimension is what determines the wavelength at which it becomes electrically significant.

Aperture Leakage in Practice

The general rule of thumb used in shielding design is that an aperture becomes a significant leakage path once its largest dimension approaches roughly one-twentieth of the wavelength at the frequency of concern — well below the point where the aperture is obviously "big." At 2.4 GHz (wavelength ≈ 125 mm), this puts the threshold at just over 6 mm, meaning even a modest ventilation slot can leak meaningfully at Wi-Fi/Bluetooth frequencies if it isn't broken up. Where ventilation is required (for thermal reasons) inside a shielded section, the standard mitigation is a honeycomb or perforated pattern of many small holes rather than one large opening — each individual hole stays well below the leakage threshold at the frequencies of concern, while the total open area for airflow can still be substantial.

EMI Gaskets

Where a soldered can isn't practical — an enclosure lid seam, a removable shielded assembly, a connector panel cutout — an EMI gasket provides the same continuous conductive path across a joint that must remain serviceable or that has real mechanical tolerance to absorb. Common gasket types include conductive elastomer strips (silicone loaded with conductive particles, offering environmental sealing alongside shielding), knitted wire mesh gaskets, and metal fingerstock strips (spring-like metal fingers that maintain contact across a range of compression). Selection depends on the compression force available from the enclosure design, whether environmental sealing (dust/moisture) is also required, and the gasket's own shielding effectiveness at the frequencies of concern — an under-compressed or corroded gasket degrades the shield at exactly the joint it was meant to protect, often without any obvious visual sign of the failure.

When a Can Is (and Isn't) the Right Fix

A shielding can is a targeted tool for a specific, localised coupling problem — not a general-purpose EMC fix. It's justified when good layout practice (see how to reduce PCB EMI for the layout-level techniques that should be applied first) has already been followed and a specific section of the board still needs isolation: a switching regulator physically close to a sensitive RF front end on the same board, a radio module's own emissions needing containment to protect an adjacent analog section, or a pre-certified module whose certification test conditions assumed a can (see antenna types for embedded wireless designs and RF PCB layout guidelines for the layout considerations that apply around the shielded and antenna sections together). Adding a can to a board with fundamentally poor return-path design or inadequate decoupling elsewhere typically doesn't fix the underlying emissions — it just changes which edge or aperture the same noise escapes through.

Design Considerations

• Design the ground via stitching before finalising the can footprint, not as an afterthought once the can vendor's footprint is dropped in — the via pattern beneath the perimeter ground pads is what actually gives the solder joint a low-impedance path to the internal ground plane at RF, and it needs to be planned into the stack-up and placement at the same time as the can footprint itself.
• Size ventilation as many small apertures, not one large one, when airflow is required through a shielded section — see the aperture-leakage discussion above for why a honeycomb or perforated pattern preserves shielding effectiveness far better than an equivalent-area single opening.
• Verify the can vendor's recommended solder and rework process, particularly for two-piece cans where the lid must be removable for rework without damaging the frame's solder joint to the board — an improperly reworked or resoldered frame is a common source of shielding that measures worse after rework than when first assembled.
• Choose gasket compression and material based on the actual enclosure tolerance stack-up, not a nominal gap dimension — a gasket specified for a gap that's actually wider under real manufacturing tolerance will be under-compressed at some units and provide inconsistent shielding across a production run.

Common Mistakes

• Treating a shielding can as a substitute for good PCB layout. A can placed over a section with poor return-path design or inadequate decoupling contains some of the problem but rarely all of it — high-frequency energy can still couple out through the can's own ground connection or through cables and connectors that necessarily penetrate the shield boundary.
• Under-sizing or omitting via stitching beneath the can's ground pads. A solder joint to a ground pad pattern with no via stitching down to the internal ground plane can still have enough impedance at RF to meaningfully degrade shielding effectiveness, even though the connection looks solid on a continuity meter at DC.
• Adding large ventilation openings without considering aperture leakage. A single large cutout for airflow or a connector, sized without checking it against the wavelength of the frequencies being contained, can leak more RF energy than the rest of the can's solid walls combined.
• Assuming an EMI gasket performs to its datasheet figures without verifying actual installed compression. A gasket's shielding effectiveness rating assumes a specific compression range from the manufacturer's test setup; an enclosure design that under- or over-compresses the gasket in the field will not achieve the rated performance, and this is easy to miss without a physical compression check on production units.

Zeus Design's PCB and product design team specifies and lays out board-level RF shielding — can footprints, ground stitching, and gasket selection — as part of complete electronics product design for products needing certified RF or EMC performance.