The freckle-size strain gauges are nearly 10,000 times more sensitive than commercially available instruments, says Al Pisano, a professor of mechanical engineering at UC Berkeley, where researchers believe they can squeeze the entire sensor into a chip no larger than single cubic millimeter.
Unlike current strain gauges, used mostly in research labs and niche testing applications, the new gauges are considerably hardier -- they don't require delicate and sophisticated housing to protect them. Instead, they bond directly to steel. And by outfitting the tiny devices with small wireless communication chips called radio-frequency identification (RFID) tags, the researchers say dozens of sensors could communicate wirelessly to help stabilize a car.
STRUM PRESSURE. High-tech gadgets are already located all over most cars -- such as the sensors that trigger the deployment of airbags (see BW Online, 9/14/04, "Change the Oil, Upgrade the Software"). Now scientists and engineers are beginning to look at ways they can get even more sensors into hard-to-reach places. That's where RFID tags, which can attach to the sensors and transmit data wirelessly, can help by allowing carmarkers to install sensors in hard-to-reach areas such as axles.
While early prototypes show promise, the Berkeley group's gauges are still very much in the research phase. In addition to working with industry partners to further miniaturize a part of the device that interprets the sensor's measurements, the researchers also are studying ways to make the gauge reliable in changing temperatures -- still a consistent problem, but "not a deal-breaker" says Babak Jamshidi, a postdoctoral student in Pisano's lab.
Here's how the gauges work: While it's undetectable to the naked eye, a piece of steel will stretch or contract microscopically when pressure is applied. The scientists bond two ends of a strand of silicon to the steel, making it taut. Then they apply a small amount of voltage to the strand, making it vibrate as if being strummed. When the steel stretches or contracts, it alters the tension on the silicon strand. Just like tuning a guitar string, this changes the frequency of the strand's vibration.
RIDING ON AIR. The sensor detects that frequency change and uses the information to conclude just how much the steel is distorting -- the key to evaluating how much pressure is being applied. The device is astonishingly precise: It can detect stretching of less than a hundred-thousandth of a percentage point.
Using that data, the sensors could enhance a "smart" suspension and correct for even the tiniest bumps in the road -- or deduce if a vehicle is about to roll long before a human could actually sense trouble.
Stability-control systems already exist, but they rely on secondary data such as how fast the wheels are spinning and the steering wheel's position. That can cause problems when the driver makes changes to the car that the computer can't anticipate, like buying different tires that change the wheels' rotation. Directly measuring force would be simpler, more reliable, and cheaper.
U-TURN. The Berkeley project, under way for three years, didn't start out to tackle car problems. Originally, Pisano and his team were trying to build a sensor to detect stress in machine gears. Their challenge was to make one small enough to fit between the teeth of a gear without being crushed. They found one of the best and simplest ways to do this was to bond the silicon directly to the steel. But the real research funding was coming out of the auto industry, so they made the switch.
So far the group is funded by the Army Research Office as well as an industrial sponsor the researchers won't name, according to Robert Azevido, another member of the lab group. In the last two weeks, the team has met with Honeywell and auto makers Toyota (TM
), BMW, and General Motors (GM
) to talk about additional funding. If the kinks get worked out, it could mean a very smooth ride down the road. Helm is a reporter for BusinessWeek Online in New York