Peak efficiency figures are little more than marketing hype

It is a well-known fact that engineers do not appreciate being advertised to. In my experience, they tend to prize function over style and want accurate information without attempts to massage the data. This is why the all-mighty datasheet continues to be the primary-source of technical information engineers turn to when evaluating components for their design. But here, as in advertising, figures are sometimes cherry-picked to shed the best possible light on a particular component. This practice is done throughout the electronics industry and, in our industry - power supplies - this comes in the form of efficiency figures, typically quoted as the highest point along the efficiency curve of a particular device. When peak efficiency is used as a determining factor to select a power supply for a particular application, the design engineer must dig deeper to understand the test conditions used to derive this number; line, load, temperature, and airflow can all affect the stated efficiency. For example, an ac-dc power supply's efficiency quoted at 230 Vac line voltage may actually operate at levels 6~10% lower if run at 120 Vac. As another example, when a power supply is tested for efficiency immediately at start-up, the number will appear higher than when measured after the unit has reached thermal equilibrium, which in reality is much closer to real world conditions. In short, peak efficiency data can provide a decent snapshot of how the power supply will perform within an application, but it does not paint the entire picture. Systems run in a variety of modes of operation and the power supply needs to be able to operate efficiently and effectively over the application's complete range of loading conditions. Because of this, OEMs also need to seriously consider the entire efficiency curve, as well as the no-load power draw of the supply. Today power supplies from the majority of vendors deliver peak efficiency percentages in the low-to-mid nineties. While this performance is certainly impressive, many applications will not operate at the power supply's efficiency "sweet spot" the majority of the time. If the power supply is not optimized to address this type of loading profile, overall efficiency will be well below the peak figure, making the touted peak power figure little more than marketing hype. Standby power As existing power topologies begin to reach their limit with regards to operating efficiency improvements, a new battle line is being drawn in the quest to reduce overall power consumption. It is estimated that anywhere from 5-10% of the total power consumed in the United States is through ‘vampire draw', power lost while electronic equipment is in standby or sleep mode. Governments have already identified that this is a major problem in external power adapters. Agencies such as the California Energy Commission (CEC) and the European Union through their EuP directive have put measures in place to set limits on the standby power consumption in all external ac-dc adapters shipped into their jurisdiction. The strictest standard, currently classified as Level V, sets the no-load limit at 0.3 W for power supplies rated under 50 W and 0.5 W for power supplies rated between 50 and 250 W. Although standby power consumption regulations do not currently exist for most internal ac-dc power supplies, greater attention is now being paid to this measure in the end equipment that incorporate these modules. CUI have been working to address this trend, most recently in the medical and home healthcare industry. The medical electronics market, currently estimated to be $150 billion in size, is expected to grow at a 9% rate or the next 5 years. As new medical equipment proliferates into the market, greater attention will be paid by customers and regulators to the energy consumption of these designs. In response to this trend, CUI has recently released a line of open frame ac-dc medical power supplies that not only provide operating efficiencies over 91%, but also offer standby power levels as low as 0.3 W. The VMS series is available in power levels of 20, 40, 60, and 100 W, and housed in compact open frame packages. Flattening the curve It's clear to many within the industry that simply driving towards higher peak efficiencies in power supplies in not the right answer, and in order to achieve greater real-world efficiencies, we need to flatten a supply's efficiency curve. We're never going to reach a situation where a supply delivers 100% efficiency across the curve, but power supply manufacturers throughout the world are looking fervently for ways to create a device with the peak maximized across a significantly broader range of the spectrum. Here at CUI, we're applying greater levels of intelligence to achieve this. For example, our NEB series and NQB series intermediate bus dc-dc converters incorporate a 32-bit ARM-based microcontroller running power-optimizing firmware. This provides peak performance across a much wider range of loading conditions when compared with the vast majority of competing devices currently on the market. The efficiency curves below highlight the NQB series next to a more conventional dc-dc converter. The NQB series exhibits a very flat curve from 20% all the way to 100% load, maintaining performance close to peak efficiency throughout this range. Conclusion Efficiency is an essential metric when choosing a power supply, but engineers need to consider more than just the peak figures and instead examine the complete curve in relation to their line and load profile to truly improve the power efficiency performance of their designs. As an industry, we need to make it easier for engineers to select the power supply that is right for their design. Peak efficiency figures provide an easy way of comparing competing power modules, but it's too simplistic and has turned into little more than a marketing figure. Instead, we need to evaluate how we present the complete picture in a standard, yet easy to interpret way. CUI