During 7 B.C., Greek soldiers—or hoplites--carried a large, concave, circular shield called the hoplon. Made from bronze and wood, the hoplon or Argive shield protected hoplites from chin to knee and—when overlapped—became the basis for the Greek phalanx.
At the beginning of the 20th century, Ethiopians used their shields when fighting against an invading Italian army. The constant banging on the shields by the Ethiopians mingled with war cries as a terrifying prelude to battle. Ancient armies deployed shields to protect from spurious attacks and to form war strategy. PCB design often relies on different types of shields to protect sensitive circuits from stray RF signals.
Within those circuit outposts, renegade wires and traces do everything possible to allow electromagnetic interference (EMI) to escape into the outer reaches of their domains. Those skirmishes never end. Shielding should exist as part of your overall PCB design strategy. Rather than viewing shields as an add-on feature, shields and shielding designs should become part of the initial design process for PCBs.
What Can PCB Shielding do for You?
RF front end, switching, and clock circuits emit EMI through conducted paths or through radiation. A combination of good PCB design and good shielding mitigates EMI. Good PCB design for EMI revolves around the layout, the placement of filters, and ground planes. A well-designed PCB minimizes parasitic capacitance and ground loops.
Instead of a providing a physical barrier against spears and swords, PCB shielding simulates a Faraday cage connected to ground. While the ground plane makes up the bottom side of the cage, you can use metal shielding to form the other five sides. Metal, magnetic materials, and gaskets used as shielding prevent EMI emissions.
A shield causes reflection loss and absorption loss of electromagnetic energy. Reflection loss varies with the type of field and occurs as the air-to-shield or shield-to-air loss. As the field propagates through the shield, absorption loss occurs. Absorption loss varies with the type of shielding material. Shielding effectiveness equals the sum of reflection and absorption losses. The conductive barrier cloaks all or part of the circuit, absorbs and reflects radiation from loop antennas, and establishes electrical isolation.
While your design may not need fort Knox, a chassis can be helpful.
The concept of a shield wall became famous when Roman legions used the testudo formation. You can also use shielding to form a wall that protects sensitive circuits from harmful EMI emitted by other circuits. Electromagnetic fields contain electric fields and magnetic fields that have a 90o orientation to one another.
If the electric field interacts with the magnetic field, propagation occurs. A shield wall separates sections of the PCB by attenuating the electric field (E-field) and magnetic field (H-field) of the radiating electromagnetic interference wave with a combination of metallic and magnetic shielding.
Round, or Square: Different Shields for Different PCB Needs
Unlike the hoplons, shields used on PCBs do not protect from chin to knee. Hidden enemies such as aperture radiation and cavity resonance can make life difficult. Electromagnetic interference can pop up anywhere. With electronic devices becoming smaller, shield design emphasizes a combination of lightweight, mechanical stability, and electrical efficiency.
The shielding that you select for your PCB design depends on whether the emissions occur as near-field or far-field emissions, the frequency of the interference, and whether the interference occurs as electric field or magnetic field interference. Each of these factors influences the selection of shielding type and thickness.
Most emissions under a shield occur as near-field emissions. With near-field emissions, reflection and absorption loss vary with frequency. Digital circuits cause electric field emissions that have higher reflection loss at lower frequencies and higher absorption loss at higher frequencies. Switch-mode power supplies cause magnetic field emissions with small reflection and absorption losses at low frequencies.
Protecting from loss within and external to your design will keep it lasting through any war.
Near-field and far-field emissions have different source characteristics and different E/H ratios. EMI is a far-field emission. Far-field sources have greater reflection loss at lower frequencies. Higher frequencies cause greater absorption loss.
You should consider two key variables when deciding on the shield material type. Conductivity measures the ability of the shield material to conduct electric current. Permeability measures the ability of the material to support the formation of a magnetic field within the material. A shield that has high permeability has low reluctance and can guide a magnetic field.
As with current and resistance, the lowest amount of reluctance creates a path for the magnetic field. Shielding material thickness becomes an important factor for PCBs that have low-frequency magnetic field applications.
Still Need to Breathe: Apertures Control Shielding Effectiveness
Ancient forts always had vulnerable points that an invading army could exploit. Good shielding design also considers mechanical strength and the need for air flow. Shields include apertures to reduce thermal buildup by allowing air to flow through the shield openings. The number and size of apertures in a shield design controls the effectiveness of the shield.
While many apertures reduce effectiveness, spacing between the apertures and frequency determine the amount of reduction. In addition, the amount of leakage from an aperture depends on the maximum linear dimension of the opening.
Whether you’re looking to supply an army, or just protect your PCB, your first step should always be in choosing the right PCB design software (hint - CircuitStudio is great).
[Thanks to Altium for permission to reproduce this post]
.