Issue link: https://iconnect007.uberflip.com/i/1496727
18 DESIGN007 MAGAZINE I APRIL 2023 with that. I believe it's a misun- derstanding about the the physics and mathematics involved. Simu- lating how it all works together requires an in-depth study and understanding of various Maxwell equations. But many people don't want to take the time to learn Maxwell, because he's not an easy read. At RF speeds, there's no room for error. What sort of problems do designers face in the RF arena? e problems designers will face specifi- cally with RF are numerous, and they can't be ignored. With a basic PCB, we have a much broader "window of operation and flexibility." With RF, everything gets raised to a whole new level. I fully agree there is no room for errors, and understanding the details of the circuit is probably the first hurdle a designer must con- sider. Small items on a "basic" PCB become important issues quickly, and significantly impact the design. But specifically to answer the question, one of the biggest problems is the massive amounts of energy flowing throughout the PCB that you must control. Having a lack of understanding or plan of how you intend to manage that energy is a failure at the offset. Many of today's devices have multiple RF points—Wi-Fi, Bluetooth, GPS, and cellular. How can designers ensure that these signals "play nicely" together? When I was raising my children, whenever they would become unruly, I used a common prac- tice utilized by parents all over the world: divide and conquer. I would separate them into their specific areas—and it's much the same with RF. First, always identify an RF design by its dis- tinct design features. For example, with vari- ous RF points, the same method is used: divide and conquer. Identify the various circuits and the level that they are operating, and then isolate them. at is usu- ally done by using a scaled-down version of a Faraday cage, which is an enclosure used to block or control electromagnetic fields in a PCB design. ese are cages that are grounded to control the emit- ted radiation. Placing a grounded metal box around the circuitry will isolate those EMI signals inside that area. When we bring up the discussion of Faraday cages, we must also discuss the physics involved with controlling such a massive amount of energy, and "skin effect." A Faraday cage works because of the skin effect and depth. e phys- ics of current flow through a material is much higher on the surface and decays considerably more profoundly into the material. at can occur with a metal shield box and is very com- mon in a PCB trace. at is such a common practice that the edge of the PCB is plated. Without plating, the edge of the PCB can become a weak point in the design. So these signals are entirely isolated into their specific areas, and everybody co-exists. One significant mistake oen made in RF design is inadequate grounding. A shielding configuration such as PCB Faraday cage is just half the battle. Placing enough grounding and stitching them together to keep them at the same reference point, and not allowing stray currents in the grounding plane, helps to isolate the energy into a specific area. at is a com- mon feature you will find in RF design: plentiful ground stitching to the ground plane, which is also connected to the Faraday cage. Another point to consider is how the signals get routed into another area. e key is that the specific grounding on each side of the iso- lation is independent. For example, a standard method is to use an opto-isolator. Each side of that opto will have a different and isolated ground to allow a "clean" return path for each circuit. John Watson