A Proper Approach to Shielding
In product development it is usually most beneficial from a cost, schedule, quality and performance standpoint to carefully consider and implement proper design as early as possible in the project development cycle. Add-ons and other “quick” fixes implemented later in the project are more often than not non-ideal solutions functionally, are of inferior quality and reliability, and are more costly than if they had been implemented sooner in the process. A lack of forethought in the early design stages of the project usually results in late shipments and potentially unhappy customers (both internal and external). This problem applies to any design, whether it be analog, digital, electrical, or mechanical, etc.
The cost of the shielding increases the further away it is applied from individual ICs or small areas on a PCB. Compared with shielding of individual ICs and small areas of a PCB, it costs roughly 10x to shield an entire PCB, 100x to shield a complete product, and 1000x to shield and entire assembly or compartment. The cost is really astronomical if shielding of an entire room or building is required because improper shielding (or no shielding) was implemented at lower levels.
A “nested” shielding approach is a possible solution. A nested approach is one where shielding is applied at each of the lowest possible levels of a product design. For example, shielding is first applied to:
- Individual ICs/small area of the PCB, followed by
- Entire PCBs, then
- Sub-assemblies, and finally
- To complete products.
A nested shielding approach is one that results in the lowest overall cost to manufacture a quality product, on time, and within performance specifications.
Shielding at the Lowest Possible Levels
Shielding at the lowest possible levels (individual ICs, small areas of the PCB, and the PCB level), makes a lot of sense for several reasons:
- Enclosure shielding does not help attenuate interference between individual ICs located on a PCB whereas, PCB level shielding does help attenuate interference between individual ICs.
- From a practical/cost-efficiency level, typical enclosure shielding technology is incapable of providing significant attenuation performance at higher (GHz) frequencies, whereas PCB level shielding does provide this performance.
- Cost and weight of shielding at higher levels is minimized through effective use of shielding at the PCB level.
- From a susceptibility stand-point, modern ICs with their ever-shrinking silicon features, faster rise-times, and lower noise margins, can be made to function dependably in the noisy atmosphere that they are often required to operate in, simply by employing shielding at the PCB level.
- Integration of intentionally noisy wireless communication modules within products can cause harmful inference to other sensitive analog and digital components located in close proximity. This noise can also be mitigated through use of PCB level shielding.
- Enclosure shielding is often compromised to a point of total ineffectiveness due to the need to have holes and slots added for the penetration of input /output cables, displays, ventilation, access to removal media, etc. This situation becomes less of a problem when PCB level shielding is utilized.
- Effective enclosure shielding usually requires substantial filtering of all cables which pass in and out of the product, right at the point where they penetrate the enclosure shield. It’s possible to lessen the need for this extra filtering when PCB level shielding is utilized.
Whether you design a cell phone, tablet, portable computer, or some other form of electronic product, good PCB layout in addition to PCB level shielding is critical to keeping EMI to a minimum. Ground (return) and power planes can be utilized as EMI shields of high-threat noisy signals and this technique is a good first step towards minimizing noise from these high-threat signals. One problem with this approach is that RF energy can still radiate off component leads and packages and a more complete solution is required. This is where a PCB level shield (a.k.a. “a shielding can”) can be utilized to attenuate the noise emanating from these noisy devices.
In order to provide the most benefit, a PCB level shield must form a complete six-sided metallic enclosure. This is accomplished by soldering the shield to a solid ground plane which lies underneath all the components that require shielding. To be most effective, the ground plane must not have any substantial slots or openings in it. The real-world performance of all shielding and ground planes is always compromised by apertures such as holes for adjustments, indicators, wires, construction seams and the gaps between a shielding can’s ground plane connections, so whenever possible these items should be avoided.
The goal of an EMI shield is to create a Faraday cage around the enclosed RF noisy components using the six sides of a metallic box. The top five sides are created using a shielding cover or metal can, while the bottom side is achieved by using the ground plane within the PCB. In an ideal enclosure, no emissions would enter or exit the box. Unwanted emissions from these shields does occur, such as from holes perforated into soldered cans that allow thermal heat transfer during solder reflow. These leaks can also occur from imperfections along an EMI gasket or solder attachments. Noise can also escape from the spaces between ground via-holes used to electrically connect the shielding cover to the ground plane.
PCB shields are traditionally attached to the PCB using through-hole solder tails, manually soldered after the main assembly process. This is a time-consuming and costly process. If maintenance is required during setup and servicing, access to circuitry and components under the shields requires de-soldering. In densely populated PCB areas containing highly sensitive components, there is a high risk of expensive damage. There are manufacturers of shield cans that provide solutions which mitigate these problems.
The typical attributes of PCB level shield cans are as follows:
- Small footprints;
- Low-profile configurations;
- Two-piece design (fence and cover);
- Through-hole or surface mount;
- Multi-cavity patterns (isolate multiple components using the same shield);
- Virtually limitless design flexibility;
- Ventilation holes;
- Removable covers for quick access to components;
- I/O holes;
- Connector cutouts;
- Enhanced shielding with RF absorbers;
- ESD protection with insulator padding;
- Reliable protection from shock and vibration using secure locking features between the frame and cover.
Typical Shielding Materials
A wide range of materials are generally available for shielding, including brass, nickel silver and stainless steel. The most common types are:
- Tin plated cold rolled steel (cheapest option)
- Tim plated copper
- Nickel silver
- Stainless steel
- Tin plated phosphorous bronze
In general, tin plated steel is the best choice for shielding below 100 MHz while tin plated copper is best above 200 MHz. Tin plating allows for the best soldering efficiency possible. Because aluminum on its own is not easily soldered to a ground plane with its heat-sinking properties, it is not generally used for PCB level shielding.
Depending on the regulatory burden of the end-product, all materials used for shielding may need to be RoHs compliant. In addition, if a product is intended for hot and humid environments, galvanic corrosion and oxidation may be of concern. If in doubt, check suitability of the shielding material with the supplier.
References and Further Reading
- In Compliance Magazine. (2018, August 2). What Every Electronics Engineer Needs to Know About: Shielding.
- Armstrong, K., EMC Design Techniques for Electronic Engineers, Armstrong/Nutwood UK publication, 2010.
- Armstrong, K., EMC for Printed Circuit Boards – Basic and Advanced Design & Layout Techniques, Armstrong/Nutwood UK publication, 2010.
