Since the 1960s, when Switch Mode Power Supplies (SMPS) first emerged, there have been numerous technological advancements that have captivated power engineers. In the 1970s, magnetic integration, soft switching, MOSFETs, and digital control were all groundbreaking innovations that pushed the boundaries of what was possible in power design. These technologies weren’t just about improving efficiency—they were about achieving more power from smaller, more compact solutions.
From a broader perspective, though, the development of SMPS seemed to plateau after the 1980s. What I mean by “smoothness†is that it became quite predictable and, at times, even a bit dull. So whenever someone claims they’ve made a revolutionary breakthrough in power electronics, my instinct is to ask, “What’s new under the sun?†It's not that I don't believe in progress—it's more about the rarity of truly novel ideas in this field.
Figure 1: The 1970s were full of exciting developments, but things have been relatively quiet since then.
When I started my career as a power designer in 1999, I believed the industry would keep pushing toward higher power density. I had tools like increasing switching frequency to shrink transformers and inductors, using soft switching to reduce losses, and better MOSFETs to lower conduction losses. All the textbooks I read reinforced this mindset.
For years, I aimed to build a MHz-level isolated power supply, no matter the application. But one day, an experienced designer shared his research with me, stating that from an efficiency and cost standpoint, 500kHz was the practical limit for commercial isolated power supplies.
It seems to hold true. For example, the LM5025A and UCC2897A can operate up to 1MHz, and they’ve been popular in telecom DC-DC converters for 50–200W. Yet, I haven’t seen many designs using them above 500kHz. Similarly, in AC-DC applications, even the latest analog PFC controller, the UCC28180, is designed to run below 250kHz. Popular topologies like phase-shifted full-bridge (PSFB) and LLC are widely used in servers and TVs, but I haven’t come across any products using controllers like the UCC28950 or UCC25600 at frequencies above 500kHz.
Why? The controller is rarely the bottleneck. The real limiting factor is the quality factor (FOM) of the MOSFET. Other challenges include increased core loss in magnetic components at MHz frequencies and parasitic inductance in capacitors. However, MOSFETs have long been the key constraint.
Over the past decade, GaN FETs have introduced a real breakthrough. My team has worked closely with GaN manufacturers to develop gate drivers like the LM5113 and UCC27611. With a significantly improved FOM compared to traditional MOSFETs, GaN has reignited my passion for designing power supplies operating at MHz or even 10MHz.
Driven by the rapid growth of wireless and wireline communication equipment, the power density of isolated DC-DC modules reached new heights in 2013—such as a one-fourth brick module reaching 864W. By 2014, the industry was moving toward 1kW. In the next 1–2 years, MOSFETs, hard-switching below MHz, and digital control will continue to play a major role.
I've seen the strong desire among isolated power supply designers to push beyond MHz. I’m optimistic that within 3–4 years, we’ll see actual MHz products on the market. To make this happen, the industry must fully understand how to leverage GaN technology. Passive component suppliers need to accelerate their efforts to support MHz operations. And power supply designers must develop topologies that can handle ultra-high frequencies without sacrificing switching performance. When all these elements align, the market for isolated power supplies will undergo a dramatic transformation.
30L Agriculture Drone,Fertilizer Drone,Gfrp Agricultural Sprayer,Fold Sprayer Drone
Xuzhou Jitian Intelligent Equipment Co. Ltd , https://www.jitianintelligent.com