Printed circuit boards (PCBs) implement their functions using three main disciplines. Signal integrity (SI) allows visibility of how signals will behave as they traverse circuit board traces, vias, connectors, and passive components, while power integrity (PI) studies a design’s power delivery networks to determine aspects like power plane impedance and decoupling capacitance. Instances of electromagnetic interference (EMI) occur when an external source generates a disturbance that affects an electrical circuit via electromagnetic induction, electrostatic coupling, or conduction.

Historically, SI, PI, and EMI have always been treated as three separate issues, despite how SI has attained near perfect power supply, while PI hasn’t properly factored the impact of switching signals. Many industry experts believe both signal and power integrity are functionally synonymous, and should be simultaneously simulated to avoid over-optimistic results. As device speeds increase, new issues among SI, PI, and EMI become more critical, leading many industry experts to believe we’ve reached a point where engineers can’t design high-speed digital systems without clearly understanding the relationship between the three disciplines.

Many industry experts are speculating on what the future has in store for the three disciplines. One pundit is Steve Sandler, managing director at Picotest, who moderated a panel of seasoned engineers in a keynote event at DesignCon earlier this month. During this session, panelists gave their predictions on what the next five years holds for EMI, PI, and SI. Throughout his career, Sandler hasn’t shied away from making bold industry predictions, and has been spot-on about quite a few noteworthy projections over the years.

“I’ve been making predictions about technology for about as long as I’ve been an engineer,” Sandler said during introductions at the keynote panel.

Sandler’s first prediction regarded the future of computers. He described the first PC he ever had, a model that (by today’s standards) would be considered obsolete. Sandler believed PCs would eventually become ubiquitous, and turned out to be correct about that particular prediction.

“32 years later, my younger daughter asked about one for her birthday. We built her one, the LED blinked, and my reputation held up, so I was pleased, and it proves the PC is still there,” Sandler recalled.

In 1977, Sandler worked on rendezvous docking displays for the space shuttle, taking in serial data and putting out coding data for seven-second displays using 13 boards, which he predicted could one day be done with six. About 12 years later when the first DC/DC brick converter was introduced, Sandler predicted the bright future this technology had, and shares the same sentiments for the future of Gallium Nitride (GaN). Mr. Sandler hasn’t refrained from predicting what the future may hold for SI, PI, and EMI technologies.

Sandler is among the industry experts who think the three disciplines are functionally synonymous, and not recognizing this particular attribute has and will continue to be problematic.

“There’s this fundamental issue, where these three groups are trying to stay segregated. It’s difficult to combine them because they don’t even have a common language; the jargons and buzz words are different for the same things,” Sandler said when I sat down with him after the keynote.

This all circles back to Sandler’s big prediction on the three disciplines, which he believes should all be simulated simultaneously when being tested. Sandler talked about the first simulation company he worked for, which conducted simulations of large satellite systems. One of the first things Sandler said he established with the company were the guidelines and disclaimers about what particular services the company conducted. This was important for the company since most of their projects had multi-billion dollar budgets, which left virtually no room for error, so being distinct about their services to customers was essential. Sandler said one of the very first things he established at the company were guidelines and disclaimers. Sandler said every one of these reports contained the disclaimer that their simulations reports don’t include circuit board effects.

“I’d say flash forward 20 years and I would say ‘don’t bother’ because it’s all printed circuit board effects. You look at projects that have not one wireless transceiver, but three or four—like phones, all in one box,” Sandler said. “How do I tell if I have an EMI, PI, or SI issue? They’re all tangled together on an itty bitty little board, where everything talks to everything else.”

Regarding his prediction of synchronizing the three PCB disciplines, Sandler believes one of the trends we’re beginning to see are certain industry experts, which he described as the “smart guys,” beginning to realize that we can’t keep putting their synchronization off forever because the systems are polluting each other.

“Where’s it going to start? It’s not going to start with the guys at the bottom; it’s going to start with the people that are willing to put in the effort and already understand this technology. Some of these experts are still trying to figure out how they separate PI, SI, and EMI. I would say at the end of the day you don’t have to, because you’re still trying to make it three things when it’s just one,” Sandler added. “What we have to do is figure out what is the language we’re going to unify these three things into one thing? Because whether or not I have a power supply problem or an SI or PI problem, they’re all going to come out of the system the same way.”