Courtesy of Shad Roundy
Upon graduation, Shad Roundy hopes to land a university professorship. He says that as well as enjoying teaching, he appreciates the flexibility academia offers to define his own research goals. (Click for larger
image.)
|
Mechanical Engineering
graduate student Shad Roundy is power hungry. Literally.
Roundy is in need of electricity for PicoRadios, tiny transceivers
on a chip in development at the Berkeley Wireless Research Center
(BWRC). Outfitted with pinhead-size micromechanical sensors and
sprinkled in the hundreds throughout a building, these cubic centimeter
devices could, for example, keep a constant vigil on light, heat
and airflow and wirelessly relay that data from node to node until
it reaches a central computer. Real-time environmental information
could then enable the creation of numerous "microclimates" of
varying degrees of temperature and light, making the structure
more comfortable for its inhabitants while drastically reducing
power consumption.
But what powers the PicoRadios? Wiring is expensive and often
impossible, batteries are impractical for long-term applications,
and solar is less than ideal in dim office buildings. Thatís why
Roundy decided to harvest power directly from the building itself,
in the form of the ambient vibrations present inside any "living"
structure.
Vibrations are all around us, Roundy explains. Sometimes we
notice them - shaking windows caused by a passing truck. Most
often we don't - the hum of computer monitors, the continuous
shudder of heating and cooling ducts. All of this subtle kinetic
energy can be harnessed and converted into electrical energy.
Not a lot, but enough to keep the PicoRadios talking.
As
the cantilevered mass bounces up and down, power is generated
in this prototype of a piezoelectric converter. (Click for
larger image.)
|
"The idea for this came from self-winding watches," says Roundy,
who chose to focus his PhD thesis on this form of micropower because
it's largely uncharted research territory. "We were first looking
at random human movement to generate power. But many buildings
have false ceilings filled with pipes and ducts and these are
also good sources of vibration."
With mechanical engineering professor Paul K. Wright and BWRC
scientific co-director Jan Rabaey, Roundy is designing two devices
that harvest low-level vibrations. While the latest generation
of vibrational energy scavengers are the size of quarters, Roundy
plans to demonstrate much smaller versions within a year.
The first device is an electrostatic converter that employs
a tiny variable capacitor for storing energy. The converter contains
two parallel plates, one on springs. As one plate moves back and
forth, voltage in the capacitor is cyclically increased and released.
This mechanism, Roundy says, is well-suited to be manufactured
en masse and on the cheap the same way silicon microchips are
fabricated. But the design drawback is that the capacitor must
be zapped with power once before the device will work on its own.
Capacitive
converters like this one can be shrunk down to the millimeter
scale much the same way microchips are etched into silicon.
(Click for larger image.)
|
The second approach, likely to boast higher power density, is
based on classic piezoelectric conversion, where mechanical stress
on certain materials generates voltage. Roundy's piezoelectric
generator resembles a two-layer diving board with a boulder on
the end of it. When the device is shaken, the cantilevered diving
board-like beam bends, creating tension in the top layer and compression
in the bottom. The opposite occurs when the beam bends back up,
enabling power to be drawn from both motions.
Roundy's first prototypes draw 70 to 80 microwatts of power
sitting on an operational microwave oven. While this is 400 times
less energy than it takes to make an average household light bulb
glow, it's not too far below what the current PicoRadios demand.
"I'm coming from the bottom and they're coming from the top
and hopefully we'll meet somewhere around 100 microwatts," he
says. "After all, we're not talking about powering cellular phones
or laptops. These mechanisms are for devices that only need to
transmit a few meters within ubiquitous networks. Vibrations really
are only useful for a niche group of applications. But that niche
is certainly there."
Energy
Scavenging Overview
Shad Roundy's
home page
PicoRadio
Berkeley Wireless Research
Center