Exemplary illustration of the measured influence of the RAI flag on the energy consumption of an NB-IoT module when transmitting different payload lengths.

© Fraunhofer IIS

Energy harvesting offers the possibility of using energy available from the direct environment of a device as electrical energy to supply electronics. Consumption optimization is essential to enable the permanent, self-sufficient operation of sensor nodes with the low electrical energy that can be generated by energy harvesting (EH). This makes it possible to recharge batteries during use or dispense with them completely, but which energy harvesting technologies are suitable for use?

Everyone knows the most widely used technology for energy harvesting: solar cells that convert ambient light into electrical energy. Typical power yields of a solar cell with an area of 1 cm2 for indoor operation, for example, are 12 µW/cm2 at an illuminance of 500 lux, which corresponds to typical office lighting. Outdoors, several milliwatts per square centimeter of solar cell can be expected, depending on the illuminance, about 2 mW/cm2 at 50,0000 lux, which corresponds to direct sunlight.

At present, thermoelectric converters - so-called thermogenerators that use temperature differences to generate electrical energy - are still not widely used when it comes to making sensors energy self-sufficient. However, they can be used very effectively for sensors on supply pipes, for example, which have a temperature difference to their surroundings due to the flowing medium. Motors, machines or gearbox bearings that become warm during operation, or even cooling units, are suitable for this energy generation. A temperature difference of just 3 to 4 Kelvin is sufficient to provide power in the range of 100 µW via Peltier elements and suitable voltage converters. One side of the generator is connected to the temperature source, for example a hot water pipe.

Utilizing vibrations and deformation

The NB-IoT sensor node developed at Fraunhofer IIS is supplied with energy via the heat of the thermogenerator and thus records the consumption data on a supply pipe of a heating system.

© Fraunhofer IIS

Mechanical energy converters that use vibrations or deformation for this purpose are another option for the self-sufficient energy supply of sensors. Energy converters made of piezoelectric materials are used here, which react to deformation by generating electrical charges. If a coil is combined with a movably mounted magnet, vibrations and relative movements of motors, pumps or containers can be converted into electrical power. The typical yield is between a few 100 µW and a few milliwatts and scales strongly with the vibration amplitude used, the vibration frequency and the size and weight of the energy harvesting generator used.

Clever power management

A key component for energy harvesting is power management in the form of voltage converters and charging circuits. They adapt the voltages obtained to the system's energy storage and consumers. The circuits must be highly efficient in order to utilize as much of the harvested energy as possible. They must also be able to operate at very low currents and voltages in order to be able to work even with minimal amounts of energy. Current voltage converters work from as little as 20 mV in order to utilize the low voltages of a thermogenerator - perfectly suited for use with small temperature differences. Other converters can rectify and down-convert the amplitudes of up to 100 V of piezo converters at the lowest currents with high efficiency.

The future of self-sufficient sensors

Output power of a piezoelectric generator as a function of frequency and load resistance.

© Fraunhofer IIS

Powering wireless systems from energy harvesting sources opens up a wide range of new areas of application and therefore data sources for Industry 4.0 applications and the Internet of Things. As the power loss of microelectronic circuits continues to be reduced, more and more use cases for wireless sensors can be operated permanently with energy harvesting. In addition to power management, system integration is also essential when implementing fully autonomous sensor systems. Here, the power supply, radio and sensor technology must be optimally coordinated and optimized according to the use case in order to achieve the desired performance with minimum size and low costs. Compared to other LPWAN systems, NB-IoT requires significantly more energy. A supply from energy harvesting sources is therefore much more difficult to implement and remains the subject of current research activities.

In this context, the supply of IoT devices from energy harvesting sources is currently also being considered in standardization at 3GPP and IEEE, for example, with regard to protocol optimizations under the keywords 'Zero Energy Communications' or 'Ambient IoT'. The supply from radio field energy using radio frequency energy harvesting, similar to RFID transponders, is also being discussed.

Regardless of the use case, the selection of the wireless system and its parameterization is also decisive for the energy consumption of the overall system. In the case of autonomous operation from energy harvesting sources, this is crucial for the reliable functioning of the system. The advantages and disadvantages of data preprocessing directly at the node and the event-dependent reduction of the amount of data to be transmitted must be considered and the overall system must be coordinated with regard to energy optimization.
At Fraunhofer IIS, energy self-sufficient IoT systems with energy harvesting and efficient wireless communication are current research topics and the subject of public projects as well as feasibility studies and developments for partners. Use cases for environmental monitoring and consumption data acquisition with solar, vibration and thermal harvesting have already been implemented.