Estimated reading time: 6 minutes

Elementary, Mr. Watson: High Power: When Physics Becomes Real

Have you ever noticed how high-speed design and signal integrity classes are always packed to standing room only, but just down the hall, the session on power electronics has plenty of empty chairs? It's not just a coincidence; it's a trend I've observed over the years as both an attendee and instructor.

High-speed topics carry a certain glamour. Engineers are eager to hear the latest about PCIe, DDR5, SerDes, or 5G techniques because we can find these technologies in the newest phones, laptops, or cars. They feel cutting-edge, fast-moving, and immediately relevant.

By contrast, when you look at the seminars on power—on the buses, converters, insulation rules, and thermal management that actually make those high-speed systems possible—the turnout is often lower. Power integrity doesn't generate the same buzz as signal integrity. I have often referred to high power as the plumbing of PCB design: It’s unseen when it works, but it’s catastrophic when it fails. Nobody thinks about the pipes until a leak sends water down the walls; then the plumber becomes the hero.

In the same way, power systems rarely get attention when everything runs smoothly, but the moment something goes wrong—arcing, smoke, or shutdown—the power engineer is suddenly the most important person in the room. Power design is invisible by design, until it isn't.

So why does power draw less attention than the “sexier” topics in high-speed design? Partly, I blame the lack of visibility: when power fails, it doesn't produce a "cool" crash screen. Power feels unforgiving: Mistakes in signal routing cause glitches; mistakes in power design can cause catastrophic failures. Many engineers assume someone else has "already solved it" with off-the-shelf modules.

Several times in my career, I've had the pleasure (if you can call it that) of working with true high-power systems, the kind where you didn't flip a switch with your finger; you nudged it with a broom handle while wearing protective gear. Just in case things went wrong, you were tethered to a safety hook so someone could drag you away if the equipment decided to make things, let's say, interesting.

Recently, I had a firsthand look—checking off an item on my technical bucket list—of seeing just how quickly the industry is changing. Whilein San Francisco, I rode in an autonomous taxi through the city's busy streets. As it navigated entirely on its own, I was reminded of Johnny Cab from the 1990 science fiction classic “Total Recall,” with Arnold Schwarzenegger. What was once pure imagination on the movie screen had now become a reality rolling down Market Street. It was a surreal moment in my life.

It was an eye-opening experience to sit in the back seat, , watching the wheel turn itself, and feeling the car respond to traffic, pedestrians, and lights without a human driver. It underscored how far transportation has advanced in just a few decades, but more importantly, it highlighted the unseen technology making it all possible. Behind the smooth ride lies an enormous amount of power electronics—from high-voltage traction systems to the converters that feed the sensors, processors, and AI hardware that guide the vehicle.

I was acutely struck by the thought that autonomy isn't only about algorithms, cameras, and LiDAR. It depends equally on the high-power systems that supply reliable energy to everything else. Without that backbone, the car doesn't move, the processors don't compute, and the promise of autonomy collapses.

So, who will be laying out these advanced systems? As the industry shifts toward higher voltages, faster switching devices, and tighter safety standards, the pool of PCB designers fully equipped to handle high-power design is surprisingly shallow. Many have strong backgrounds in digital logic, signal integrity, and embedded systems. But far fewer have been trained to think in terms of insulation coordination, partial discharge, creepage and clearance, or field management.

Universities often emphasize coding, AI, and high-speed design, while power electronics—especially in the kilovolt, multi-kilowatt class—gets far less classroom attention. At the same time, companies are rapidly transitioning to 400 V, 800 V, and even 1,000 V automotive platforms, where the consequences of design mistakes are far more severe than a corrupted data packet.

That creates a skills gap because technology is advancing faster than the workforce pipeline. This will only worsen the skills gap in the high-power design area. What we really need are cross-disciplinary PCB designers, professionals who understand not just schematics, simulation, and high-speed design but also real-world physics, materials science, and safety standards for high power. In the words of Forrest Gump, that's all I have to say about that.

High power can be a slippery term unless placed in context. In general engineering terms, high power refers to systems that handle such large amounts of energy that thermal management, insulation coordination, and safety become primary design concerns.

However, high power is generally defined not by voltage or current, but by their product—the amount of energy being transferred. What counts as "high" depends on context. “High” can mean above 100 watts in consumer electronics, but just a few watts is considered high power in RF. High power in EVs might mean hundreds of kilowatts, or hundreds of megawatts in industrial systems. In practice, high power begins when heat must be actively managed, insulation distances must be calculated, and failures pose real safety risks. It's the point where the physics of heat, fields, and safety dominate the design, and in autonomous vehicles, it is the invisible backbone that makes autonomy possible. A design engineer, when thinking of high power, told me he means that "this is the point where physics stops being polite and starts getting real."

But actually, when you look at high power vs. high speed, the laws of physics don't change as we move from a 3.3-volt trace to an 800-volt bus. Maxwell's equations, Ohm's law, and Joule heating are exactly the same equations we learn in the first semester of electrical engineering. What does change is how dominant those effects become once the voltages and currents rise.

At low power, parasitics are background noise. A few nanohenries of inductance or a fraction of an ohm in a trace hardly matter. A slight temperature rise can be ignored. A layout shortcut rarely causes more than minor performance issues. In short, the physics is there, but it behaves politely.

But with high power, those same "small" effects become the main character in the movie. A few nanohenries of inductance can cause destructive voltage spikes. A few milliohms of resistance can translate into significant heating. A sharp copper corner or a contaminated board surface can lead to partial discharge or arcing. The physics has not changed, but the consequences have scaled up dramatically.

That's why high power is both fascinating and intimidating. It forces designers to confront the same basic rules of electricity and magnetism, but at a scale where the margins are thin, the by-products are unavoidable, and the cost of ignoring them is no longer a glitch but a failure that can take down the entire system.

This column originally appeared in the October 2025 issue of Design007 Magazine.