Reading time ( words)
We spoke with Ken Wyatt of Wyatt Technical Services during DesignCon 2021. In this interview, Ken discusses his DesignCon tutorial and how proper stackup planning can help in the fight against radiated emissions. He also points out that many PCB designers lack a true understanding of EMC and radiated emissions in general, and why some knowledge of physics is so important to today’s PCB designers.
Andy Shaughnessy: Hi, Ken. Long time, no see! You just spent three hours on your feet doing a tutorial today, right?
Ken Wyatt: Exactly. My feet hurt, but otherwise I’m doing fine.
Nolan Johnson: I heard you had an interesting tutorial topic. Why don’t you give us a quick rundown?
Wyatt: Sure. As a consultant for EMC, most of my clients have issues with radiated emissions and that’s usually the top issue. The tutorial was all about troubleshooting radiated emissions and doing pre-compliance testing on that topic and the ways that they can do the troubleshooting and pre-compliance testing in-house.
Shaughnessy: What were some of the high points of your presentation?
Wyatt: We covered some basics of EMC as far as product design, and we really concentrated a lot on the PC board design and stackup. That seems to be the heart of the problem for radiated emissions.
Johnson: Stackup is the heart of the problem?
Wyatt: Yes. Stackup is probably the number one issue as far as poor PC board design. A lot of people don’t understand how digital signals propagate in PC boards. I always thought that it was electron flow in the copper traces, and that’s not how it works in the physics of things. Digital signals are propagated as electromagnetic waves, and those waves occur between the circuit trace and the return plane and propagates through the dielectric. Most designers don’t understand that. That was a new concept today.
The stackup is important because you want to confine that electromagnetic field from source to load. Therefore, that really dictates the stackup because you want an adjacent return plane right next to every signal layer. The same is true with the power distribution network. You want an adjacent return plane adjacent to it. So, by following those two rules, it really dictates the stackup design.
Shaughnessy: You say you could use stackup to preclude or to head off the radiated emissions, right?
Wyatt: If you do the stackup right, you’re well on your way to creating a board design that is going to be relatively low EMI. We also talked about some troubleshooting techniques and product design features, like where to terminate cable shields and the importance of having a solid return plane with no gaps in it. A gap can create a lot of common mode currents throughout the board and can couple to IO cables and power cables.
Things like gaps or pigtails and cable shield terminations can cause common mode currents to flow and couple onto the outside of shielded cables, and turn them into nice transmitting antennas. We also talked about how you can set up your own pre-compliance testing using an antenna, and how you can set up an antenna in a conference room, for example, and make some measurements and compare those to what is measured in the compliance test chambers.
Shaughnessy: You were saying before that there is a definite lack of educational resources in radiation. Why is that?
Wyatt: Most universities and colleges don’t teach electromagnetic compatibility, and nowadays all products must meet certain limits and be immune to certain environmental effects. There’s just not enough of us who are doing the training, so most product designers go through endless hoops and cycles back and forth to the compliance test lab, between there and trying to troubleshoot fixes.
Shaughnessy: Is this your first trade show since COVID?
Wyatt: Yes, this is the first face-to-face one we’ve had since the pandemic hit.
Johnson: Back when I was a software engineer, I was working in integrated circuit design tools where I worked on parasitics extraction: estimating the parasitic resistances capacitances between geometries in the design. We’re now getting down to dimensions on the PCB that are very close to ICs back then.
Wyatt: That’s right.
Johnson: Are we starting to have parasitic effects? Is that something that needs to be factored into the analysis for the circuit boards?
Wyatt: Well, it’s a good question. Parasitic effects have always been an issue with EMC because there are internal board couplings that you may or may not want. Either capacitive or inductive coupling, depending on how you route traces.
Johnson: As current runs down a trace, you get a magnetic field, and if you have other traces within range of that magnetic field, now you have parasitic effects. Your point earlier about needing to pay attention to stackup, and the dielectric, and those other details becomes critical to minimizing the effect of those electromagnetic fields on your design.
Wyatt: Well, it’s very true. Those trace-to-trace couplings, that’s crosstalk. One way to reduce that issue is through partitioning of different circuit functions on the circuit board. Keep the analog stuff away from the digital, keep the DC-to-DC converters away from wireless modules and analog circuits. As long as you try to partition by keeping noisy circuits away from quieter circuits, that’s going to help a lot in the area of noise coupling.
Johnson: That’s where you get into floor planning.
Shaughnessy: People have been moving away from floor plans, though, but then there’s a trend back toward it.
Wyatt: Well, probably the most important takeaway is how signals propagate in circuit boards. It’s not through electron movement in copper; it’s through electromagnetic waves traveling through the dielectric and trapped between the circuit trace and return plane. The circuit trace is actually a waveguide. It’s guiding that wave and keeping it away from quiet stuff.
Johnson: I hadn’t thought of it in that regard before, but that’s exactly right.
Shaughnessy: That’s interesting. You really have to understand the basic physics to understand how to fight radiated emissions.
Wyatt: I didn’t understand the basic physics until maybe five years ago when I had multiple conversations with Dr. Eric Bogatin, and took some classes from people like Dan Beeker, and Rick Hartley. We all keep learning and trying to pass it on.
Shaughnessy: That’s fascinating.
Wyatt: It is. It’s been a fascinating field for me. Thanks so much.
Shaughnessy: Thank you.