RF Shielding Can PCB Design & Mounting Guide

What Is an RF Shielding Can on a PCB?

An RF shielding can is a metallic enclosure soldered or clipped onto a PCB to contain electromagnetic emissions from a noisy circuit or protect a sensitive receiver from external interference. A properly mounted rf shielding can pcb assembly delivers 30–60 dB of isolation from low-MHz through millimeter-wave frequencies, depending on material, seam quality, and grounding. The can works by forming a near-continuous Faraday cage around the target circuitry, with the PCB ground plane serving as the sixth wall of the enclosure.

RF shielding cans are standard practice in smartphones, Wi-Fi modules, GPS receivers, automotive radar front-ends, and any RF PCB where multiple radio sections share a single board. Without them, coupling between a power amplifier and an adjacent LNA can degrade sensitivity by 10 dB or more—enough to halve a link budget.

Types of RF Shielding Can PCB Configurations

One-Piece Stamped Cans

A single drawn or stamped metal box, typically 0.15–0.20 mm (6–8 mil) thick tin-plated steel or nickel silver. One-piece cans are the lowest-cost option and are soldered directly to a pad fence on the PCB. Their drawback: once soldered, rework requires desoldering the entire can, risking damage to components underneath.

Two-Piece Frame-and-Lid (Fence-and-Cap)

A perimeter fence is reflow-soldered to the board first, then a press-fit or clip-on lid snaps into place after component inspection and testing. This is the dominant rf shielding can pcb topology in high-volume consumer electronics because it allows optical inspection, ICT probing, and component rework before the lid goes on. The RF shield frame-and-lid assembly process is critical to achieving repeatable isolation.

Board-Level Shielding (BLS) with Compartments

A single fence with internal divider walls creates multiple shielded compartments—useful when a Bluetooth radio, a cellular PA, and a GPS LNA all sit within centimeters of each other. Each compartment can achieve independent isolation, but the internal walls must also contact the ground plane through their own pad rows.

Material Selection for RF Shielding Can PCB Assemblies

The shielding effectiveness (SE) of a metal enclosure depends on reflection loss, absorption loss, and re-reflection correction. For thin-wall cans at RF frequencies, reflection loss dominates below about 1 GHz, while absorption loss becomes significant above a few GHz.

• Tin-plated cold-rolled steel (CRS): High permeability (μ_r ≈ 200–1000) gives excellent magnetic-field shielding below 100 MHz. Cost-effective and easily solderable. Wall thickness of 0.15 mm provides roughly 40 dB absorption at 2.4 GHz.
• Nickel silver (Cu-Ni-Zn alloy): Better corrosion resistance, good solderability, lower magnetic permeability than steel. Preferred when environmental reliability matters more than low-frequency magnetic shielding.
• Mu-metal (Ni-Fe alloy): Very high permeability (μ_r up to 50 000 after annealing). Reserved for specialty applications like MRI coil PCBs or precision clocks where sub-MHz magnetic fields must be attenuated.
• Aluminum: Lightweight and high conductivity, but difficult to solder with standard Sn-Ag-Cu paste. Usually attached with conductive adhesive or mechanical clips.

For most wireless products operating between 700 MHz and 6 GHz, tin-plated CRS at 0.15–0.20 mm thickness is the cost-performance sweet spot. Consult the IPC-2221 generic PCB design standard for ground-plane and pad-fence guidelines that affect shielding integrity.

PCB Design Rules for Shielding Can Mounting

Ground Pad Fence

The can’s perimeter lands on a continuous or near-continuous row of solder pads connected to the ground plane. These pad-fence details determine whether your rf shielding can pcb layout achieves its target isolation or leaks at resonant frequencies. Key rules:

• Pad width: Match the can wall thickness plus 0.2–0.3 mm per side for solder fillet—typically 1.0–1.5 mm total pad width.
• Pad gap: Keep gaps between pads under λ/20 at the highest frequency of concern. At 5.8 GHz (λ ≈ 52 mm), that means gaps no wider than 2.6 mm. For 28 GHz 5G, gaps must shrink below 0.5 mm—effectively a continuous pad ring.
• Via stitching: Place ground vias along the pad fence at intervals of λ/20 or tighter. A 0.3 mm drill on 1.0 mm pitch is typical for sub-6 GHz designs. These vias connect the surface pad to every internal ground layer and the bottom ground plane, forming a via wall that extends the Faraday cage through the board stackup.
• Keep-out zone: No signal traces or components within 0.5 mm of the pad fence inner edge. Traces crossing under the fence wall are a common source of coupling leaks.

Signal Entry and Exit

Every trace that enters or exits the shielded area is a potential leak path. Route signal traces perpendicular to the fence wall and through a gap in the pad fence that is as narrow as possible. Place a ground via on each side of the signal trace at the crossing point. For 50 Ω microstrip, maintain the controlled-impedance geometry right up to the fence wall—do not widen or neck-down the trace.

Stackup Considerations

The PCB ground plane that forms the bottom of the shielding cavity should be a continuous, unbroken copper pour. Any split in this plane—even a narrow anti-pad around a via—degrades isolation. In a 4-layer RF stackup, layer 2 (the first internal plane) typically serves as this reference ground, with via stitching connecting it to the surface shield pads.

Estimating EMI Risk Before Adding an RF Shielding Can

Not every circuit needs a shielding can. Adding one increases cost, height, weight, and thermal resistance. Use the estimator below to input your operating frequency, source power, victim sensitivity, and board separation distance. It returns a rough coupling estimate and a risk rating that helps you decide whether an rf shielding can pcb solution, a guard trace, or simple distance is the right mitigation.

If the tool returns a coupling margin below 10 dB, a shielding can is strongly recommended. Margins of 10–20 dB may be managed with careful layout and guard traces alone. Above 20 dB, the natural path loss and ground-plane isolation are probably sufficient—but always verify with a prototype measurement, because real-world coupling through shared power rails and substrate modes can surprise you.

Soldering and Attachment Methods

Reflow Soldering (Most Common)

Shield fences and one-piece cans are placed by pick-and-place onto solder paste and reflowed alongside other SMD components. The can or fence must withstand the peak reflow temperature (typically 245–260 °C for SAC305). Tin plating on the can must be solderable and free of oxidation—boards and cans stored in humid conditions should be baked beforehand, following proper PCB baking and moisture control procedures.

Selective Soldering

Used when the can is added after other components are already soldered and cannot tolerate a second full reflow. A selective solder nozzle or miniwave applies solder to the fence pads only. This requires accessible pad geometry and is more common in mixed-technology builds.

Conductive Adhesive

Silver-filled epoxy provides a lower-temperature alternative (cure at 120–150 °C) for heat-sensitive assemblies or aluminum cans that resist soldering. Conductivity is lower than solder, so SE may drop 5–10 dB compared to a well-soldered joint. Adhesive joints also age differently—thermal cycling can crack the bond, so reliability testing per JEDEC JESD22-A104 temperature cycling is advisable.

Mechanical Clips and Spring Fingers

Press-fit lids on two-piece shields use spring fingers that contact the fence walls. The contact force and finger pitch determine the electrical continuity. Worn or bent fingers are a field-failure mode—specify a minimum contact force per finger (typically 0.05–0.15 N) and test isolation after mechanical shock and vibration.

CAN Bus Termination Inside Shielded Enclosures

In automotive and industrial systems, CAN bus transceivers often sit inside RF-shielded modules—especially in radar sensor units and telematics control units. The CAN bus differential pair exits the shield through filtered feedthroughs or common-mode chokes. Proper termination (typically 120 Ω at each end of the bus) is essential to prevent reflections that radiate as broadband EMI, partially defeating the purpose of the shield. Use the tool below to check your CAN bus termination values.

If the calculator indicates an impedance mismatch greater than 10 %, add a discrete 120 Ω termination resistor at the transceiver inside the shielded enclosure, placed as close to the CAN_H / CAN_L pins as possible. Split termination (two 60 Ω resistors with a capacitor to ground) provides additional common-mode filtering, which reduces the EMI that the shielding can must handle.

Testing Shielding Effectiveness

Shielding effectiveness is measured in decibels as the ratio of field strength without the shield to field strength with the shield in place. Common methods:

• Near-field probing: A small loop or monopole probe scans above the shielded area while the circuit inside is active. Quick and inexpensive, but results are geometry-dependent and hard to correlate to far-field SE.
• Dual-TEM-cell method: The PCB is placed between two TEM cells; one drives a known field, the other measures leakage. Provides repeatable, broadband SE data from 100 MHz to several GHz.
• IEEE 299 / IEEE 299.1: The IEEE 299.1 standard for small enclosures defines procedures for enclosures smaller than the lowest test frequency’s wavelength—directly applicable to board-level shield cans.

Target SE values depend on the application. For a Wi-Fi/Bluetooth combo module at 2.4 GHz, 40 dB is a common requirement. For a 77 GHz automotive radar front-end sharing a board with a microcontroller, 50–60 dB between the RF and digital sections is typical. Achieving these numbers requires both a good can and a good PCB layout—a perfect rf shielding can pcb installation on a poorly stitched ground plane will still leak.

Common Pitfalls and How to Avoid Them

1. Incomplete via stitching: Missing even a few vias along the fence creates slot antennas that radiate efficiently at resonant frequencies. Audit via placement in your DRC.
2. Traces routed under the fence wall: A signal trace crossing beneath the solder pad row couples energy in and out of the shielded cavity. Route all signals through designated entry points.
3. Thermal vias inside the cavity: Large thermal-relief vias under a power device can break the ground plane continuity. Use filled and capped vias, or move thermal vias away from the shield perimeter.
4. Ignoring the lid-to-fence contact: A two-piece shield is only as good as its worst spring-finger contact. Specify minimum contact force and inspect after environmental testing.
5. Forgetting height clearance: The tallest component inside the cavity plus 0.3–0.5 mm clearance sets the can height. Verify in 3D before ordering tooling.

For complex multi-shield assemblies, a turnkey RF module service that handles both PCB fabrication and shield attachment under one roof reduces the risk of misalignment between the bare board and the shielding hardware.

Frequently Asked Questions

How much isolation does an RF shielding can provide?

A well-designed and properly soldered board-level shield can typically provides 30–60 dB of isolation, depending on frequency, material, wall thickness, and the quality of the ground-plane via stitching beneath the can perimeter.

Can I reflow-solder an RF shielding can with standard SMT processes?

Yes. Tin-plated steel and nickel-silver cans are designed for SAC305 reflow at 245–260 °C peak. The can or fence is placed by pick-and-place onto solder paste and reflowed alongside other components. Ensure the can material datasheet confirms reflow compatibility.

What via pitch is needed under the shield fence?

Place ground-stitching vias at intervals no greater than λ/20 at the highest operating frequency. For 5.8 GHz, that means ≤ 2.6 mm pitch. For 28 GHz, ≤ 0.5 mm pitch. Tighter is always better.

Does adding a shielding can affect thermal performance?

Yes. The enclosed air volume has limited convective cooling. High-power components inside a shielded cavity may need thermal vias to a heatsink on the opposite board side, or the can itself may include ventilation slots (which reduce SE and must be kept below λ/20 in their longest dimension).

When should I use a two-piece shield instead of a one-piece can?

Use a two-piece frame-and-lid whenever you need post-reflow access for inspection, testing, or rework—which is nearly every production scenario. One-piece cans are reserved for very low-cost, high-volume products where rework rates are negligible and height is critical.