Crossing the Digital Divide
It wasn't that long ago that companies were debating the viability of digital versus analog approaches. Analog is highly entrenched and is unlikely to be unseated in the foreseeable future. Digital is a challenger to this throne, and speculation has varied considerably as to how much (or whether) it could displace analog. The answer to these questions has always depended on two things: price and applications. The way those two things are converging is providing a clear picture of where digital is heading.
Price has been the biggest issue so far. It led to the whole debate of whether to use a digital solution over an analog solution. If the analog solution worked perfectly well, why pay for something you didn't need? Digital ICs didn't reach price parity with analog ICs until last year, which is when we predicted sales would pick up for digital control products. That has turned out to be the case.
Our conclusion was based on an analysis of the pricing of over 250 switch-mode controller ICs (about 150 digital controllers and about 100 analog controllers) for the 42-month period between January 2004 and June 2007 from 24 IC makers. In a little over three years (early 2004 to June 2007), average prices for digital power ICs dropped from about $6 at the beginning of the period to under $3. Prices for digital controller ICs continued to drop and were under $2 by the first quarter of 2008.
We also predicted that as more "specialized" digital controller ICs were launched, they would be tailored to the specific needs of unique application segments, and the pricing variations would continue to narrow. This is where the digital price advantage is "crossing over" with the digital application advantage.
Digital was originally targeted at high-performance, higher-end applications. This is where its value was greatest, but even then, the application could often function perfectly well with an analog solution. Companies had to "make a case for digital," to prove that it was "viable." With the advent of the Centralized Control Architecture (CCA) and adaptive control, all of that is changing. It is no longer a matter of replacing an analog product with a digital product. Systems are now requiring functions that only digital can provide. Digital control opened up a whole world of possibilities for original equipment manufacturers, and they responded by making systems that must have digital power management.
Two examples of the digital wave of the future are recent product announcements from Ericsson and Powervation. The former exemplifies the large, long-standing power supply company. The latter represents the smaller, entrepreneurial power supply company. In other words, both sides of the house are coming to the market with products that veer from a simple "search for analog and replace with digital" model.
Ericsson introduced the 30A Power Block, which addresses "the emerging technology in power distribution and management called Centralized Control Architecture (CCA)." System architects are requiring suppliers to offer solutions that address multiple rails (sometimes up to 150). The CCA is a "natural evolution of the intermediate bus architecture," according to the company. The Power Block is also very much related to the control IC, which is digital. The CCA is a totally new architecture based on the requirements of digital power management in systems. Digital control is no longer simply "an option."
Whereas Ericsson's product is based on an emerging architecture, Powervation's product is based on the emerging functions that digital has always promised – and which could "kill" analog in certain applications. The PV3002 chip is the industry's "first" digital power conversion IC with adaptive control – what the company calls "Auto-Control®." It is claimed to be the first product to "reliably adapt power conversion to changing system behavior and unpredictable variations," guaranteeing "unprecedented stability over a wide range of conditions."
This is a game changer, with one company saying that Powervation's solution could eliminate functions like "tunable loops and others" used in analog point-of-load converters. Powervation claims that their new IC enables digital power management "at no extra cost" at the system level. When applications require functions that can't be provided by analog, and the digital solution is at the same price point (or less) as analog, then the digital divide has been crossed and is rapidly closing.
Some companies don't seem aware of this – they may have digital products, but they downplay their role in the broader market. They see themselves as "minor players" when up against the larger power supply companies, or they don't consider "digital" to be their main focus. This is short-sighted. "Digital" has expanded beyond controller ICs and digitally controlled power supplies. There are now advanced components and technologies that are enabling digital solutions – many of which are not specifically digital. For example, DrMOS, silicon-carbide and gallium-nitride are reaching a point in their development where they are helping digital cross the divide.
It's still premature to predict how much digital control will penetrate the power supply market. But the tables are turning, and it is now more accurate to say, "As long as analog solutions are viable…" Digital has come into its own, and the balance will be shifting significantly over the next five years.