Sensing powered by the environment

The next generation of noise and vibration sensors will power themselves, and do so without the need for wires. Advances in electronics mean that the energy derived from vibration in the environment is all that is needed to meet all power requirements. The proof-of-concept prototype is currently running round a major city attached to a car engine, but first commercial applications are envisaged in condition monitoring in industry. Professor Neil White and his colleagues at the Department of Electronics and Computer Science at the University of Southampton, have been investigating the possibility of powering sensor electronics using features of the environment they monitor. Solar cells, powered by light and thermocouples powered by heat differentials have been in use for years, but the Southampton team has been looking at ways of powering sensors by vibration. The obvious approach is to use piezeoelectrics, and the team has spent what it describes as "many years" researching this solution. However, the teams preferred approach is to use old fashioned electromagnetic induction, which it found produces more power – mW as opposed to microwatts. Resonant power One example of a piezoelectronic device produced by the team was made of 316 stainless steel, 0.1mm thick, 23mm long with a maximum width of 20mm. The beam, tapered to obtain a constant strain along its length, had a 0.8g mass attached to its free end and power was generated by coating it with lead zirconate titanate (PZT). The resonant frequency was found to be 80Hz and the optimum load resistance was found to be 333kohm. Despite the sophistication of the design, the maximum power generated by a 0.9mm displacement was found to be just 2 microwatts with maximum voltage of around 1.2V. There are ways of improving the output of piezoelectric devices, as reported in past issues of Eureka. These include turning to more advanced (and expensive) materials and methods of manufacture. However, the big breakthrough has been to turn away from latest solid-state technology and go back to the technologies discovered by Michael Faraday – electromagnetic induction. The team has investigated both a fixed coil excited by moving magnets attached to a vibrating stainless steel beam, and the converse, a fixed magnet with a moving coil. White says that the moving magnet approach is the easiest to make, with the prototype unit exhibiting two disk-shaped neodymium iron boron magnets attached to a stainless steel beam which is fixed at one end. The magnets are mounted on a 'C' shaped core and provide the mass. Arranging the magnets in this way provides a uniform magnetic field in the air gap. The main purpose of the core is to provide a path and guide the magnetic flux through it with a minimum of flux leakage. The coil is made up of a number of single, solid core enamelled copper wires. It is placed in the air gap between the magnets, at right angles to the direction of their movement. As the housing is vibrated, the mass moves relative to the housing and energy is stored in the mass/beam system. This relative sinusoidal displacement, causes the magnetic flux to cut the coil. This, in turn, induces an electromotive force on the coil due to Faradays' law. The magnitude of this voltage being proportional to the rate of change of the coil position. The electrical system is a first order LC circuit with the inductance of the coil in series with the load resistance and the parasitic resistance of the coil. For convenience and ease of fabrication, the air gap is a relatively large 3mm. The core thickness is determined by the need to carry the magnetic flux in the circuit without exceeding a nominal saturation flux density of 1.6 T. The beam in the first prototype is 0.2mm thick, 3mm wide and 10mm long with the effective mass of the magnetic core assembly being 500mg. The resonant frequency was found to be 327Hz and optimum load 0.28ohms. The device generated 0.53mW at a vibration amplitude of 25 microns. This corresponded to a maximum beam deflection of 0.94mm at the resonant frequency. Optimisation of the design was achieved with the aid of MatLab and VF-OPERA, a finite element field analysis CAD package. Since then, a test unit, measuring about 50mm long and about 25mm high, has been test driven round Southampton attached to the engine of a Volkswagen Golf. It is found to produce 3.9mW peak, and just over 1mW average. The demonstrator unit has a commercial Analog Devices chip accelerometer as sensor, and delivers its signal to the outside world through a photodiode. An obvious application for the technology is in the monitoring or turbine rotor bearings – a particularly unfriendly environment for wiring or batteries. The next stage in the development should lead to matchbox-sized devices, with natural frequencies from 100Hz to a few kHz. For wearable electronics, there should be no problem in developing devices with natural frequencies of a few Hz, should there be a market demand. There have already been a number of developments aimed at powering re-chargers for mobile phones, including some described in Eureka. These range from piezoelectric generators in the heels of shoes, to toroidal coils worn on a belt, powered by movements of a magnetised ball. Thick-film devices are likely to be the eventual solution of choice and, according to White, these could well take the form of screen printable magnets and flat coils. It may be possible to fulfil one of the goals of the original project, which was to make micron-sized generators for micron-sized systems, but with much greater efficiencies than look possible using piezoelectrics. The first stage of the research project – to develop the most promising kind of generator – was funded by the EPSRC and is now almost complete. The team is now asking for input from industry, in order to develop the most appropriate configurations for possible commercial products. Pointers It is possible to make vibration and acceleration sensors powered by the vibration of the platforms on which they are mounted Power generation using electromagnetic induction is about 1,000 times as effective as that based on piezoelectrics A prototype unit has been experimentally mounted on a car engine and monitored during drives around a UK city