Power on the Double

Mooly Eden envisions a day when he can haul his laptop out of the office for the day and not pack his power supply. He's not asking for miracles. He merely wants to use his laptop for what it was intended — an unwired extension of his office. But while new wireless technologies have allowed him to cut the umbilical Ethernet cord to the network, he's still tethered to the wired world by his power cable. Eden, vice president of Intel's Mobile Platforms Group, wants to cut that cable.

It may not seem like a very ambitious dream, but of all the technical issues belaboring the wireless industry it has become one of the most difficult problems to tackle. Battery power has been an afterthought in the wireless world, but now a sector that has been written off as commoditized is beginning to see new life as both computer and handset vendors are running into the power limitations of today's current technologies. Wireless and computing technology is moving forward at a fantastic clip, but batteries aren't able to keep up.

“If you have a car that can go 800 miles an hour, that's great, but if you can only drive for 10 minutes before it runs out of gas, it's useless,” Eden said. “It's the same with mobile computers. We can do all kinds of crazy stuff, but we dare not go outside that power envelope of three or four hours. Give us a bigger power envelope and we can do miracles.”

Though Eden is Intel's overall director of marketing, his title is a bit misleading. One of his main responsibilities is to supervise the development of Intel's mobile processors and chipsets, including its new wireless computing technology. But while Intel's desktop divisions have devoted themselves to living up to the ever-faster philosophy of Moore's Law, Eden's mobile platforms division has found itself with a lot more concerns than processing speed. With enhanced speed comes increased power drain and when a silicon radio is thrown into the mix, power consumption increases exponentially. And while processing speeds may be almost doubling every year, battery life is increasing at the lackluster pace of less than 5% each year.

So instead of just making fast chips, Eden has to make fast chips that consume as little power as possible. His team has made advances in adaptive processing technology that essentially whittle a processor down into separate parts and put those individual parts to sleep whenever they're not being used, waking them up when they're needed. It has begun integrating radios directly into the silicon chipset, optimizing the power consumption between different components. It's developing new ways to curb thermal leakage. It has tweaked its LCD screen technology to take full advantage of ambient light. A lot of these innovations are already in many of the wireless laptops, PDAs and smart devices in the market, and Intel continues to improve upon that technology. But Eden admits those efforts are not enough.

With wireless becoming almost a de facto standard technology on PDAs and laptops and data increasing its grip on the wireless phone, the current optimization technologies won't be enough to support the power demands of 3G. In fact, if the industry isn't careful, power limitations will actually hinder the proliferation of wireless data. Wireless video broadcasts will be of little use to the average consumer if his or her phone goes dead five minutes after the broadcast concludes. New applications designed to use powerful W-CDMA, CMDA EV-DO and WiMAX broadband connections won't be heartily embraced if they cause laptops to slide under the three-hour battery power mark. And one of broadband's biggest appeals, the “always on” aspect marketed so heavily during the launch of 2.5G networks, will be utterly useless.

The industry has to find away to overcome that power barrier, Eden said. And it can't be done just through optimization or improving the efficiency of current technology, he added.

“We can't just improve battery technology at a linear rate,” Eden said. “We need a huge jump ahead. We need a revolution.”

Intel and a good deal of other companies in the wireless industry are counting on that revolution coming from an entirely new kind of battery technology, replacing the current lithium ion technologies the same way the first rechargeable nickel cadmium batteries leaped over standard alkaline AA cells more than a decade ago.

Intel has been looking into the next evolution of rechargeable cells using zinc oxide, such as those being developed by Zinc Matrix Power. But most of the excitement about power sources seems to be coming from the fuel cell industry.

“The number of companies working on portable fuel cells has tripled in the last few years, and every one of the laptop manufacturers has a fuel cell component program in development,” said Jim Balcom, president and CEO of PolyFuel, a fuel cell developer working with Intel. “It's safe to say that there's an enormous amount of new interest in the industry.”

But fuel cells aren't a new concept. They've been talked about for years as alternate energy sources for everything from automobiles to defibrillators. Much of the technology's focus over the years has been on shrinking the fuel cells themselves from vaguely car-sized to more manageable volumes. Fuel cells work through a chemical reaction that converts hydrogen and oxygen into water and carbon dioxide, producing an electrical current in the process. It's cheap (hydrogen is the nature's most abundant element) and environmentally friendly (water and carbon dioxide both occur naturally). To create that reaction, though, the hydrogen and oxygen have to pass through a porous membrane where they come into contact with a catalyst. To generate any sort of significant current the membrane must have a tremendous amount of surface area. Much of the research into fuel cells has been into wrapping that surface area into compact three-dimensional constructs to reduce the fuel cell's bulk and optimize the efficiency of the reaction itself so as to avoid the necessary lugging about of inordinate amounts of hydrogen.

Several fuel cell companies have managed to shrink the current generation of methanol fuel cells strong enough to power a camcorder or PDA into a baton-sized package, and continuing R&D efforts will soon bring them down to a form factor capable of fitting into a laptop CD-ROM slot. Before the end of the decade, fuel cells will be small enough to slide into handheld devices, said Gregg Makuch, director of marketing and administration for Neah Power Systems, a fuel cell firm also with products in development with Intel.

“You are going to see the first fuel cell products out there late this year and in early 2005,” Makuch said. “At first they'll be box chargers, capable of charging up a laptop battery without a power cord. Around 2007, though, you'll see them making their way directly into devices.”

Makuch predicts there will be that initial surge in power that Intel and so many other hardware vendors crave, as much as a 200% to 300% boost in milliwatts over today's Lithium Ion batteries. What makes the technology most attractive, however, is that the cell recharges by taking more fuel. A simple plastic cartridge containing a fresh dose of methanol (an alcohol comprising the hydrogen necessary for the fuel cell reaction) is popped into the fuel cell to return it to full power. At first the methanol cartridges will cost a few bucks a pop, but eventually as fuel cells become popular and cartridges become standardized, you'll be able to buy them at gas stations and news stands for the price of a butane lighter, Makuch said.

While much of the excitement around fuel cells and other power technologies has been driven by laptops, smartphones may ultimately be the major beneficiary of the technology. Companies like Intel are focused on increasing the mobile lives of their laptops out of the office. Conversely, wireless handset vendors will become more focused on preserving the long battery life that their customers have to come to expect from their mobile phones.

In fact, power drain has already become a major issue with major smartphone and wireless PDA vendors. While companies like PalmOne and Microsoft have managed to keep their battery lives to between eight and 10 hours of talk time (a number significantly reduced if more demanding data applications are used), they've had to make significant trade-offs.

“We get bombarded with new technologies every year, and we'd love to incorporate those technologies into our devices,” said Stephen Maes, senior product line manager for handhelds at PalmOne. “Unfortunately, battery technology isn't progressing fast enough to support those technologies. Our engineers get increasingly frustrated because we can't move as fast we want to.”

The vendors have compensated by shrinking the non-power components of their devices to make more room for battery space. While the form factors of Microsoft's PocketPC's and PalmOne's Tungsten and Treo lines haven't noticeably decreased in size over the last year, the percentage of the phone's volume devoted to battery has sizably increased. PalmOne was able to upgrade features in its newest Treo 600 by simply giving it an oversized battery. And Microsoft estimates that 60% of most of its smart devices is devoted to battery space.

The power crunch will only become more severe. As 3G networks become more prevalent, power demands on handsets will increase. The single greatest power drain of any wireless device is the radio itself, and as the devices start transmitting and receiving more data over longer periods of time, that drain becomes more intense. In Japan and Korea, where 3G networks and broadband-enabled handsets are common, the problem has already come to the forefront. Mobile audio and video streaming has taken such a toll on battery life that it has created an industry around recharging dead phones on the go.

But that revolution in the handset battery is still much further off than the revolution in the laptop battery, probably well into the end of the decade. Until then the vendors will have to learn to deal with the hands they're dealt. And none of them appear ready to fold just yet.

“If the battery companies could give us a 50% increase overnight, we'd stand up and cheer,” said John Starkweather, product manager of Microsoft's mobile and embedded devices division. “But since we're faced with such a slow rate of growth, those are the parameters we're working in. We've already made some significant improvements in power efficiency. As an industry we should be proud of that. We can't just point the blame at the battery companies. We have to be smarter ourselves.”

ON THE WEB

For more information on alternative power solutions for wireless devices, go to our Web site:
WWW.WIRELESSREVIEW.COM