As industry readies itself for the increased application of self-powered wireless sensors Drake Calvin examines the latest in piezoelectric innovation.

Sensor technology is big business and with energy prices on the rise it is easy to understand why researchers are quenching industry's thirst for energy-harvesting solutions. With this in mind, more and more energy sources are being explored – piezoelectric energy-harvesting applications using vibrations from motion and wind have long been in development, and now, as large-scale hydro-generation looks to move downstream, out of our rivers and our estuaries and further out into our open seas, it is hardly surprising that researchers are examining the small-scale energy-harvesting potential of another stage of the water cycle – rainfall.

Articles in the popular press lauded the innovative rain energy research coming out of the Atomic Energy Commission (CEA) in Grenoble, France, and surmised that piezoelectric materials, which convert mechanical strain into electrical energy, could supplement large-scale solar power by generating energy both night and day. This, the press says, will mitigate the efficiency problems that continue to suppress solar power's widespread commercial application.

The tagline to this press supposition is that Jean-Jacques Chaillout and the CEA research team will first prove the raindrop concept in wireless industrial sensors that will detect and report limescale build-up, which reduces the efficiency of nuclear power station cooling towers. And then comes the solar revolution.

Chaillout confirms his droplet-powered limescale sensor is unique; he knows of no prior work to harvest the mechanical force of raindrops, however, he plays down the press' idea of twinning it with solar devices: “The efficiency difference is too high to exploit them both together,” he says.

Over-egging the piezoelectric pudding has become something of a norm; many innovative devices are heralded as solutions for large-scale application - only the truth often turns out differently - the opportunities remain interesting but seriously narrowed. Now, let's examine Chaillout's research; when he thinks it will come to fruition; and if the application of raindrop power in sensors is still an innovative step too far.

Rain research

Writing in Smart Materials and Structures, Chaillout and his team set out to discover whether raindrops could be harvested for energy and if the energy available could be exploited to power a wireless sensor.

In the theory portion of their research the team concludes that effective energy recovery from raindrops requires a piezoelectric material that is very thin, not pre-stressed, and that has a slightly smaller diameter than the maximum diameter of the impacting drop. The team classified droplets into two categories: drizzle raindrops (1mm in diameter) and larger downpour raindrops (5mm in diameter).

Drizzle drops have an impact energy of 2 µJ whereas downpour drops have an impact energy of 1 mJ. Computer simulations indicate that 25 µm thick piezoelectric materials most-efficiently recover energy across a range of raindrop sizes. Despite this optimisation, the system in practice  only makes around one microwatt of power available from the smallest drops, which Chaillout says is enough to power the RF transmission of one digital bit for 10m: “Our system can be used for low rate transmission,” he says. 

This is quite low. What is the principle limitation of Chaillout's device? And what can be done to improve it?

“The principle limitation is the energy quantity of a raindrop, followed by the low efficiency of piezoelectric materials.” He says this low efficiency doesn’t matter as the team is not interested in exploiting rain and solar combined: “We will exploit our system in an industrial environment with no light but where the number and energy of the drops is higher. In this nuclear chimney environment operators need to measure the lime evolution and our system can power the sensor to do that.”

Chaillout is looking for a collaborative partner to help prove the device in an industrial setting: “It's a new way to power sensors in special environments.” If the collaboration begins it will take about two years to build the first industrial prototype.”

The team will now take the time to improve the efficiency of their system and build the prototype.

Piezoelectric concerns

The idea is certainly innovative. Wireless sensors are revolutionising the industrial landscape - allowing operators to collect data from devices that we’re previously either technically or financially impossible to wire up. Such innovation will improve both plant safety and maintenance. However, despite these advantages, Roy Freeland, ceo of Southampton University spin-out Perpetuum, casts doubt on the viability of piezoelectric sensors.

“The problem with energy harvesting is that you can produce a small amount of power relatively easily, the problem is producing enough power for it to be useful.” Freeland says that piezoelectric devices are not always the most practical choice for energy-harvesting sensors.

As discussed above, the energy available from CEA’s harvester is enough to transmit a digital bit of information across 10 meters. Freeland compares it to the vibration-energy harvesting devices they build at Perpetuum: “Using magnets and coils, we're sending 10kb of information 100 to 200m.”

“Our researchers examined piezoelectrics but felt that for practical energy harvesting a coil and a magnet is the way to go.”

So, what does Freeland see as the main limits of piezoelectrics when it comes to industrial sensors?

“Piezoelectric devices wear out quickly,” he says. “You may have a material that survives 200 million cycles, which sounds a lot, but at 50hz vibration that’s only 2 to 3 months, then the material fractures and becomes useless.”

This is in direct opposition to one of the key selling points of industrial, batteryless, wireless sensors: they can be left in isolation and won't need regular, costly, battery replacement or maintenance. Piezoelectric devices do not currently tick this box.

Even if the materials used to build piezoelectric devices become more robust, Freeland does not predict a sea change in energy harvesting: “The materials have to be improved a great deal and I still think they will never rival a coil and magnet in the sense that it's completely reliable and it's already possible to achieve close to maximum energy transduction efficiency.”

Challenging the concept

And this is the nub of Freeland’s point: he believes that energy harvesting concepts should be challenged. “The concept may work,” he says, “and it may generate electricity, but is it the most sensible and efficient way of doing the job?”

“It is quite easy to harvest small amounts of energy from almost anything: walking across a spongy floor, or attaching a knee brace to your leg, but neither are the most practical methods available.”

“If you want a human to produce electricity from motion then you’re better off giving them a bicycle with a dynamo. Strapping something to someone's leg and charging a mobile phone may sound great but it’s an awful lot easier to give them a crank handle device that turns a small generator. These devices get a lot of attention because they sound so innovative.”

So, he disagrees with CEA’s rainpower concept?

“Not entirely. In a cooling tower you have a consistent flow of water and a massive volume. It’s logical. But having said that I wouldn't be surprised if it is easier to collect water higher up the cooling tower and channel it through a paddle wheel that generates power for a wireless sensor.”

What the future holds

Freeland preaches a cautious approach and he may be right to do so. The US Defence Advanced Research Projects Agency (DARPA) funded the development of a piezoelectric device that would generate energy by harvesting the impact energy of a marching soldier's boots to power personnel equipment. Efforts to innovate such a system were abandoned due to impracticality.

Wireless sensors will certainly find mass application, they are already being used to monitor assets, resource pipelines, and train bearings – cheap and continuous process monitoring, plant safety, and pre-emptive maintenance will all follow en masse. It seems however, that whilst piezoelectric devices are certainly innovative they require further optimisation (and less over-statement) before they are considered commercially viable in this field.