International Rectifier Corp. (IRF ) isn't as famous as Intel Corp. (INTC ), but the company has been in the microelectronics game far longer. Founded in 1947, it has focused from the start on how smart power management can make machines and electronic devices more efficient. Today, IR is a world leader in the design and manufacture of chipsets that feed electricity to motors, lights, and a variety of other high-tech gear.
Chief Executive Alex Lidow is fiercely dedicated to efficiency. It's a conviction that he traces back to the 1970s when he was a PhD student in applied physics at Stanford University. He and his colleagues were inspired by new ways to think about energy use. And that focus is just as timely today, with energy prices soaring and the whole world clamoring for cost-saving, power-conserving air conditioners, computers, and other gadgets.
Costly energy has stoked demand for IR's solutions. Sales of devices for energy-efficient appliances and lighting systems both surged by nearly 60% in the year through March. And IR's power management solutions have helped hybrid vehicles take off. Consensus forecasts for the year though June predict revenues will grow by 11%, to $1.2 billion, with profits up by 93%, to $173 million. Lidow recently met with Industries Editor Adam Aston in New York to share his take on efficiency:
You are evangelical about IR's role in improving efficiency. What's to be gained?
It breaks down pretty easily. In 2001 the global energy budget was 404 quadrillion BTUs [British thermal units]. Electricity accounts for 39% of that, or about 160 quadrillion BTUs. Of that, about half goes into electric motion, not including transportation. It's things like refrigerators, washing machines, pumps, valves, conveyor belts, elevators, you name it. And 85% of those motors are electro-mechanically actuated -- either they're on or they're off. They're dumb, wasteful motors.
If we could convert all of those motors to variable speed, we could cut their [power] consumption by half, which amounts to 10% of total global energy consumption. Another fifth of all energy goes to lights -- you could save half of that easily by switching to high-efficiency fluorescents. In transportation, which eats up about 21% of the total global energy budget, if we could convert all vehicles to hybrids, you'd save another 10% of the world's energy.
So what's the total you see as possible?
Altogether, about 30% of our global energy budget can be saved. This has profound implications. If you extrapolate economic growth out to 2025, we'll need to add about that much energy to the total pie just to maintain our standard of living. Or think of it another way: 30% energy savings amounts to about $1.7 trillion a year in 2025 that we won't have spent on energy, if we deliver the savings for free.
How do "smarter" electric motors help?
Imagine if you drove your car with your foot all the way on the gas, then your foot all the way on the brake. You'd get really crappy mileage, and your car wouldn't last as long. Many electric motors work this way: either off or full-speed on. So the motors in most refrigerators, for example, control electricity with a switch -- you can hear it go on and off. When it's on, a little motor turns furiously to drive a compressor to cool the refrigerator, then shuts off completely when the temperature falls. That's not efficient. A motor drive that turns exactly as much as needed is a better solution. It goes a little faster after you open the door, to lower the temperature, then a little slower after it has been closed a while.
By varying the speed, a lot of good things happen. For one, you don't have all the energy losses from acceleration and deceleration. Also, you don't need as big a compressor, since it doesn't have to deliver all the energy in such a short period. You can also replace the old-style AC induction motor. These are about 60% to 65% efficient, but a DC brushless or permanent magnetic AC motor is 85% to 90% efficient. Put it all together, and you can save half the energy.
Yet sales of variable speed motors make up a small share of the market. Why?
You have to do it cost-effectively. IR's goal is to add this sort of efficiency for no additional cost. About 10 years ago we committed to delivering a far more complex motor drive, able to do variable speed motion, for the same cost as the technology then in use -- about $29. We built a variable motor drive for a washing machine and it added $250 to the final cost. Though expensive, it still sold -- proving there's a demand for these systems, especially in Asia where apartments are small and quiet appliances are preferred.
But how do you get down to $29?
We set out a roadmap. It involved many technologies, some of which we had, some we didn't. Developing and acquiring them defined a new strategy for IR -- what I call "technology pull." As a company, we didn't want to ride a single technology till it died, like buggy whips. Instead we wanted to have the ability to pull in the technology to follow this roadmap all the way to $29. Lo and behold, in 2004 we had a $35 to $40 solution, and by the end of next year, it will hit $29.
In the meantime, those washing machines have become the hottest sellers in the market, whether it's Maytag's (MYG ) Neptune, Miele's, or Bosch's. Sales of high-efficiency washing machines have grown about 33% a year. IR has an Intel-like market share in that business, supplying the motor drives that make them work.
How does this technology transfer to hybrid vehicles?
We make electronics that help control the electric drivetrain in hybrids. It's related to the architecture we use in washing machines and other appliances. The same "technology pull" approach is guiding our hybrids strategy. In the late 90s we asked, can we save 60% of the energy in a vehicle with no compromise in cost or performance? The answer's yes. By 2015 or so, there will be no price premium for hybrids, and they'll outperform their gasoline predecessors.
How do you lower the cost of hybrids?
The key is understanding the cost tradeoffs between old and new technologies. If you add an electric motor, you shrink the engine and use the cost savings to finance more electric systems. That's your energy equation. Today you can build a hybrid drivetrain that's about one-third electric and two-thirds gas. For a midsize car, that means you've saved about $150 by shrinking the engine. But to equal that capacity, the electric motor and batteries still cost $300 to $400. That differential is one reason hybrids still cost more.
We're not there yet, but we can see how to get to $150, especially as hybrids multiply: volume production will lower costs, you standardize some of the systems, and amortize the research and development. By 2015 or so, I think about 25% to 35% of vehicles will be hybrids.
You said that savings from better lighting still offer huge potential energy gains. This surprises me.
About 11 billion incandescent bulbs are sold every year. Each one is an enormously inefficient way to convert electrons into photons. They give off so much heat that you increase the air conditioning load. There are two good replacements: today, fluorescents, which use about one-fourth the energy of incandescents; later, possibly light-emitting diodes (LEDs).
Fluorescents cost less to run, but they're expensive.
Compared with a 100-watt bulb used for 6,000 hours, a fluorescent bulb will save you about $60 worth of energy if power costs 10 cents per kilowatt hour. Yet there are only 400 million or so fluorescents sold every year. IR sells integrated circuits that dramatically reduce the cost of controlling fluorescent bulbs and that capture all of the efficiency gains. But this is very difficult. It's important that we try to lower the price of flourescents because still today, if someone is faced with a $4 fluorescent and a $1 light bulb -- even if the fluorescent saves them $30 -- they'll often buy the $1 bulb.
You also sell power management chips to computer makers. Do you agree that info tech sales have passed their boom time?
Not at all. The broadband revolution is going to be a major transformation, on par with the birth of the PC and the rise of the Net. I think we're about to hit a significant inflection point in the next two years as various technologies -- the most promising of which is WiMax -- hit the market.
How does widespread broadband translate into better sales for you?
For consumers, the shift to broadband digital media will lead to a huge migration to new digital TVs and digital audio. There are 1.6 billion TV sets in the world, and there are about 700 million PCs in the world. So you have more than twice the number of TVs as PCs. In the next five to seven years, all those TVs will begin to go from analog to digital -- and each digital TV has about the same chip content as a PC. So you won't just see a computer upgrade cycle, you'll see a force twice as big as computers driving TVs and audio. In addition, more bandwidth directly increases [sales of] servers, routers, and switches. It won't make our job any easier. Cyclicality won't go away. But it's sure to increase the underlying growth rate.