Energy harvesting is by no means a new idea. The first hydroelectric plant which combined water and gravity to drive electricity generating turbines was built in 1882 and offered a relatively “green” and sustainable source of electric power on a very large scale. However, as this type of power source is greatly dependent on the natural terrain, large and expensive transmission networks are required. Since transmission losses rise with distance, this dramatically reduces the amount of available power. Nevertheless, in many instances only a few milliwatts of power are needed to power a wireless sensor node, so a much smaller scale solution is needed.
A much more cost-effective and electrically efficient solution is to keep the power source very close to the load creating a point-of-load design that eliminates transmission losses. However, in order to create these designs, there needs to be a readily available power source that can operate in remote areas, is cost effective and is self-sustaining, thereby requiring no servicing over many years.
The solution for these applications has re-introduced the concept of energy harvesting from a very different perspective, creating an emerging market for compact, predominantly wireless applications at the very low end of the power spectrum. These applications require output power that ranges from a few nanowatts to tens of milliwatts. Although non-traditional power sources such as solar cells (photovoltaic cells), thermoelectric generators (TEGs), thermopiles and piezoelectric transducers are known sources of electrical power, harnessing power from these sources has been challenging. Each of these require some type of power conversion circuit that can efficiently collect, manage and convert these alternative power sources into a more usable form of electrical energy to power sensors, microcontrollers and wireless transceivers. Whether the source voltage is very low and must be up-converted to be useful, or even rectified and then down-converted in some cases, specific energy harvesting circuits are necessary. Historically, these circuits have needed very complex discrete circuits with upwards of 30 components and yet still struggle to provide high enough efficiency to be of practical use. It is only recently that specialized energy harvesting power ICs have been introduced that can offer compact, simple and very efficient power conversion and management solutions.
These ultralow power solutions can be used in a wide array of wireless systems, including transportation infrastructure, medical devices, tire pressure sensing, industrial sensing, building automation and asset tracking. These systems generally spend the majority of their operational lives in standby mode asleep requiring only a handful of uW. When awakened, a sensor measures parameters such as pressure, temperature or mechanical deflection and then transmits this data to a remote control system wirelessly. The entire measurement, processing and transmission time is usually only tens of milliseconds, but may require hundreds of mW of power for this brief period. Since the duty cycles of these applications are low, the average power that must be harvested can also be relatively low. The power source could simply be a battery. However, the battery will have to be recharged by some means or eventually be replaced. In many of these applications, the cost of physically replacing the battery makes it unfeasible. This makes an ambient energy source a more attractive alternative.
Emerging nanopower wireless sensor applications
In the case of building automation, systems such as occupancy sensors, thermostats and light switches can eliminate the power or control wiring normally required and use a mechanical or energy harvesting system instead. This alternative approach can also mitigate the costs of routine maintenance normally associated with wired systems in addition to eliminating the need for wiring to be installed in the first place, or for regular battery replacement in wireless applications.
Similarly, a wireless network utilizing an energy harvesting technique can link any number of sensors together in a building to reduce heating, ventilation & air conditioning (HVAC) and lighting costs by turning off power to non-essential areas when the building has no occupants.
A typical solar energy scavenging system represented by the five main circuit system blocks shown in Figure 1 consists of a free energy source such as a small photovoltaic cell exposed to either direct sunlight or even indoor lighting. These photovoltaic cells are capable of generating over 50mW of electrical power per square cm of area, in peak sunlight and up to 100uW of electrical power in indoor lighting. However, the electrical energy they generate must be collected in a very specific manner using an energy harvesting circuit (see the second block in Figure 1) that can efficiently collect this low voltage energy and convert it into a more usable form, which can be used to continually charge a storage device. As the power generated by the solar cell will vary dramatically with the ambient lighting conditions, a rechargeable storage device such as a battery or supercap (block three in Figure 1) is required to provide continuous power when the ambient light is no longer available. In turn, the storage device, whether a battery or a supercap, combined with a simple step-down DC/DC converter (fourth block in Figure 1, which is usually not needed) can power downstream electronics while it is continually recharged. The downstream electronics will usually consist of some kind of sensor(s), analog-to-digital converter, ultralow power microcontroller and wireless transceiver (fifth block in Figure 1). These components take the harvested energy, now in the form of a regulated power supply, and wake up a sensor to take a readings or a measurements, making this data available for transmission via an ultralow power wireless transceiver. The most recent generation of ultralow power wireless microcontrollers include multiple ADCs and an integrated wireless transceiver. They generally require current levels of 20mA to 35mA for periods as short as 1mS while measuring and transmitting, after which they go into a sleep mode requiring only 3.5uA of supply current minimizing the average power requirements.
Figure 1: The five main blocks of a typical energy-scavenging system
Each circuit block in this chain has had a unique set of constraints that have impaired its commercial viability until recently. Although low cost and low power sensors and microcontrollers have been available for some time, only recently have ultralow power transceivers been integrated with microcontrollers to offer very low power wireless connectivity. Nevertheless, the laggard in this chain has been the energy harvesting IC.
Existing implementations of the energy harvester block are a relatively low performance discrete configuration, usually consisting of 30 or more components. These designs have low conversion efficiency and high quiescent currents. Both of these deficiencies result in the requirement for larger, more expensive batteries and solar cells compromising the performance of the end system. Without these larger storage elements, the low conversion efficiency will increase the amount of time required to power up a system, which in turn increases the time interval between taking a sensor reading and transmitting this data. High quiescent currents in the power conversion circuitry can severely limit the amount of “useable” energy that can be harvested and made available to the application load. Achieving both low quiescent current operation and high power conversion efficiency also requires a high degree of analog switchmode power supply expertise – which is rarely readily available.
The “missing link,” if you will, has been a highly integrated DC/DC converter that can harvest and manage surplus energy from extremely low power sources. However, that has all changed.
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.