Electronic Design

Power> Overview

New Technologies Arise To Solve Power Challenges

In both good and bad economic times, certain factors will always remain steadfast. In the digital realm, Moore's Law has been more or less a constant for the past few decades. In the power electronics realm, a host of trends follow from Moore's Law. As CMOS chips go to finer lithography with each new generation, their multiplying transistors require lower and lower voltages and higher and higher currents. These trends have driven up power demands on pc boards and put constant pressure on power-supply and power-system developers to increase the efficiency and power or current density of their supplies. At the same time, the trends toward lower voltages and higher currents have encouraged migration from centralized to distributed power architectures.

The progress associated with Moore's Law is fairly steady, creating the effect that newer chips are regularly being introduced with lower supply-voltage requirements. Rather than supplant the existing chips and their higher supply voltages, the new chips tend to simply add another supply voltage to the chip user's pc boards. This situation has greatly complicated power-system and power-supply design. For instance, more power-conversion blocks are needed on the boards, assuming a distributed power scheme. Sequencing all of the different supplies then becomes a power-management nightmare. When you combine requirements for greater power/current density, greater efficiency, and lower voltages with the need for more voltages and more difficult control, you get big power design challenges.

One of the silver linings of the recent downturn may have been that power technology companies were able to focus more of their engineering efforts on R&D. The development efforts made by these companies over the past few years has now begun to yield impressive results in a number of areas. Consider just a few of the recent board-level power developments that promise to shape power-system design in the coming year:

  • The spread of the intermediate voltage bus architecture, which has spurred innovative developments in isolated and nonisolated dc-dc converters.
  • Vicor's Factorized Power Architecture, which seeks to redraw distributed power schemes just about everywhere.
  • Digital control, which is slowly changing the way bricks, SIPs, and VRMs are designed.
  • The use of IC-style packages to build complete or nearly complete point-of-load power converters.
  • Highly integrated power-supply controllers that tempt system designers to roll their own nonisolated and isolated converters.

In portable applications, some of the recent semiconductor developments provide equipment designers with the extra firepower needed to add functionality. Meanwhile, many of the newer power-management chips are critical to the development of emerging applications like Power over Ethernet, lighting management, and energy metering.

Perhaps some of the most interesting developments are occurring at the system level. For instance, lead-acid batteries—both the guarantor and the Achilles heel of power backup—are now being challenged by new battery chemistries, fuel cells, and flywheels. These new energy storage technologies could make a tremendous impact on system reliability and energy usage for years to come. For many of these technologies, 2004 could be a landmark year, one in which they gain a foothold in the market or are surpassed by other advances.

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.