Preston Manning $100,000 Principal Award sponsored by TransAlta Corporation: Leung Thermal Accelerometer
When you stop to think about it, everything is in motion.
Your heart beats. The Earth turns. Even mountains imperceptibly rise into the sky and erode into the sea.
If everything is moving, it's useful to know how fast you and the world around you is going.
Imagine if your car had side airbags that inflated the instant they detected your vehicle was about to roll over. Or if your golf club told you how far you'd hit the ball. Imagine a sensor so sensitive, it could weigh a single living cell, yet go crashing into the surface of Mars to detect a volcano erupting.
These scenarios are all now possible, thanks to Dr. Albert Leung's revolutionary motion-detecting sensor, a heat-based or "thermal accelerometer."
Leung got the idea for his accelerometer in the early 1980s, while travelling to a job interview that led to his position as engineering sciences professor at Simon Fraser University in Burnaby, B.C.
He had been thinking a lot about micro-machining - the technology used to make and assemble the integrated electronic microchips that go into computers, amplifiers and chemical/biological sensors. But Leung was more interested in using micro-machining in an inertial sensor - a device that would accurately measure mechanical properties such as pressure and velocity or speed.
"I was on the plane and it took off, and it activated my thinking about acceleration," he recalls. "I took a piece of paper and started scratching down some different ideas, and I kept those two pages."
Conventional mechanical inertial sensors use a "proof mass," a pre-measured weight that's attached to some type of a spring to measure velocity. Acceleration produces a force on the mass - a change in velocity - and this bends or stretches the spring, providing a measurement of the acceleration.
For these mechanical accelerometers to be as sensitive and accurate as possible, the mass either has to be increased or the spring's tension has to be decreased. But this adds to the complexity of the device and to the manufacturing costs. It also makes the "mass-and-spring" accelerometer fairly easy to break if it undergoes a shock - like dropping it on a concrete floor. This sudden deceleration is measured in units of gravitational forces or g's. They are the same kind of g's an astronaut experiences inside a rocket blasting off, or that you would experience if your car rear-ended another car.
"My device is quite different," Leung says. "The sensitivity doesn't depend on a proof mass. Even if we have an extremely large over-range signal (a physical shock), up to tens of thousands of g's, it will not break."
There are no moving parts inside Leung's accelerometer to dislodge or shatter. Unlike a mechanical device, his invention measures acceleration by using heated air molecules moving within a microchip.
His tiny sensor is built from the same kind of silicon microchip that powers everything from computers to CD players to music amplifiers. Leung, using micro-machining techniques at Simon Fraser University's "clean room" laboratories, installed a miniature heater suspended on a microscopic-sized bridge inside a microchip. The heater warms the air within a tiny cavity that's etched or carved out of the silicon substrate - the base of the microchip.
Silicon is a good heat conductor that normally would disperse any heat produced by the microscopic-sized heater. Leung and his research team had to devise new micro-machining techniques to etch or remove the silicon from underneath the heater. "So you would be able to heat the air instead of heating the silicon."
To better understand the principle used in his accelerometer, check out a typical carpenter's level. The "bubble" of heated air inside his accelerometer is similar to the air bubble trapped in a liquid inside a tube on the level. The air bubble has a lower density - it's lighter than the liquid. So when you tilt the level, the air bubble rises - or accelerates against the force of gravity - within the higher-density liquid. It's the same principle you see when air bubbles rise to the top of a liquid in a glass.
In Leung's accelerometer, the heated air bubble beneath the microscopic heater has a lower density than the surrounding, cooler air within the microchip's etched-out cavity. Any kind of motion "disturbs this position of this hot air bubble slightly," Leung explains. By using two tiny temperature sensors to measure the air bubble moving inside the sensor, "we can infer the acceleration or the change in velocity . . . we're able to tell the direction and magnitude of the applied acceleration to the sensor."
Along with the inherent ruggedness of Leung's accelerometer - no moving parts to break or dislodge - his device can be manufactured for about five times less cost than conventional mechanical inertial sensors. Any standard integrated electronic circuitry manufacturing plant can mass-produce the base silicon "wafer" upon which his accelerometer is built.
"I essentially can go to some foundry that will be able to make the wafer," Leung says. "It makes it much cheaper than the ordinary sensor and also much more reliable," because microchip production must be done according to consistent, high-quality standards.
John Dewey Jones, director of the engineering science faculty at Simon Fraser University, says the Leung accelerometer's "radically new design offers greatly reduced cost, combined with sensitivities one or two orders of magnitude beyond prior technology."
Simon Fraser has licenced the technology to MEMSIC Inc., a subsidiary of leading sensor-maker Analog Devices Inc. in the U.S. In just six months since MEMSIC announced the product's availability, more than 250 customers have inquired about preliminary designs for more than 25 different applications. "Moreover, we have booked initial production orders from several customers," says Thomas Cunneen, MEMSIC's vice-president of marketing and sales.
Dynastream Innovations Inc., a "smart devices" company in Cochrane just west of Calgary, has developed the first commercial application based on Leung's technology. Dynastream, which has a three-year agreement with Nike Inc., has put an accelerometer sensor system in Nike's line of SDMtriax 100 athletic shoes. The system, used with a wristband display, instantly tells runners and walkers precisely how fast they're moving and how far they've gone.
The market for inertial sensors, currently US$1 billion, is projected to exceed $1.5 billion in 2003. In the automotive industry alone, where accelerometers are the "trigger" for airbag release systems, the market is worth US$100 million.
Leung's accelerometer is targeted for a wide variety of applications, including vehicle airbags and fail-safe systems, medical and industrial sensors, computer game "joysticks" and portable computers, home appliances, earthquake sensing, and space probes to Mars and beyond.
"MEMSIC anticipates the ability to capture 30 per cent-plus of the projected sensor market with this revolutionary product technology," Cunneen says.
"I always want to put it in a golf ball or a golf club," Leung says with a chuckle. "Then I can say to myself, 'Albert, your ball just went 357 yards!'"
The Ernest C. Manning Innovation Awards Foundation
Each year, Manning Innovation Awards presents $135,000 in prize money, distributed among four leading Canadian innovators, as well as $20,000 among eight Canada-Wide Science Fair winners. During the past two decades, the Foundation has awarded $2.75 million to encourage and recognize Canadian innovators.
Media contacts (photos available):
Dr. Albert Leung
OR: MEMSIC Inc.
Donald Park, Executive Director
Ernest C. Manning Awards Foundation