Be it through the smart energy management or air quality monitoring, smart IoT systems can significantly improve the environmental impact of a building to comply with battery regulations for IoT devices. It’s therefore no surprise the smart building sector is growing rapidly (forecast CAGR of 28.5% from 2024 to 2030), and a huge array of IoT sensors and actuators have been developed to enable ever greater levels of in-building automation, including energy management.
According to analysts, for the smart-building ecosystem the majority of this growth comes from security, facilities and energy management technologies, but there is also a notable increase in home automation that has been attributed to the rise in working from home following the Covid pandemic.
There is, however, a problem with the way these devices are powered, with a large proportion of these using disposable batteries. This causes significant problems further down the line.
For facility managers, the use of batteries creates a significant operational expense, with the cost to replace the batteries often being as high as the cost of the original device. The man hours needed to check and replace the devices’ batteries when required is also significant versus the original capital expense.
The environmental concerns related to using batteries is also making both organizations and individuals seek out alternatives. Indeed, consumer research from Simon-Kucher & Partners, as well as from Cisco suggests people are willing to spend significantly more for sustainable products and services, with them typically accepting a 20 to 25 percent premium. This premium also exists for B2B purchases, with Deloitte research showing 1.7 times more likely to pay a premium and 2.7 times more likely to make long-term commitments to suppliers with sustainable product options
Governments looking at ways to reduce battery disposal
Despite regulations on the disposal of batteries, a large proportion end up in landfill. Academic papers and multiple news media have even suggested this figure is as high as 95%.
If we look just at waste from the IoT sector, an EU government-funded report estimates 78 million batteries from IoT devices will be discarded in 2025. It’s therefore not surprising that the EU is looking at how to reduce this, and has called on IoT developers to ensure battery life outlasts the IoT devices they power.
Its report stresses the need to harvest available energy from the environment to make batteries last longer, while also reducing IoT device energy consumption.
The request for batteries to outlast the IoT devices can be achieved in three ways:
1) by engineering devices to use energy harvesting which removes the need for battery replacement altogether; or
2) using larger and longer-lasting batteries. This doesn’t get around the environmental problems when the device reaches end of life and could be seen as cynical; or
3) more cynically still, make no changes to the design but get around these recommendations by instead reducing the stated IoT device target life.
If developers take the second (and especially if they take the third) option, it is likely that legislation will eventually replace voluntary aims.
How to implement energy harvesting for the Ambient IoT
The need to implement an Ambient IoT is now crucial. While not every application will be suited to harvested power, a growing number are.
Recent years have seen improvements in the range of energy harvesting technologies (light, thermal, piezoelectric, RF, electromagnetic induction etc) and the efficiency with which they can harvest energy from their environment. There have also been significant advances in board components that help applications work with tight power budgets. These include MCUs, wireless protocols (eg LoRaWAN and Bluetooth LE) and crucially power management ICs (PMICs) that have been developed specifically for low power IoT applications.
As we’ve seen, CapEx is also a crucial factor in the buying decision. And here it should be noted that the cost of developing and manufacturing Ambient IoT systems is reducing quickly. Indeed, the technology is now at the point where Ambient IoT systems can be implemented not just within the premium that customers will pay, but also more cost effectively than some battery-based systems.
A typical ambient-IoT system comprises a raw energy harvester connected to a PMIC and energy storage, with sensor/actuators, a low-power MCU to process data, manage peripherals, and execute operations efficiently at the edge. These are typically combined with data transmission (via BLE, LoRa, Mioty, cellular or Zigbee, proprietary analog radio). Additionally, IoT systems will implement supporting elements such as memory, crystal oscillators for clocking, and security features.
All components come optimized for ultra-low-power operation and tailored to the application to ensure maximum efficiency. This includes the PMIC, which work better when it is tailored to specific behavior of the energy source, according to which harvesters is used. For example, e-peas’ range of Ambient Energy Manager (AEM) PMICs have been developed for specific energy sources and this can easily be seen in the e-peas selector guide.
IoT system developers are used to designing for tight power budgets, but these are based on the large “closed” energy reservoirs of disposable batteries. This “closed energy reservoir is sized to survive for the entire duration of the device life time. Whereas for energy-harvesting-based “open” systems this requirement must be considered in a different way : storage element is an energy buffer to supply the system when raw energy source is not availableIt is charged when raw energy source is available.
To enable this to happen it is critical to understand what energy sources are available and practical and therefore the first step is to undertake a power budget analysis to determine how the storage element can be downsized. With recent advances in PMICs (see below) allowing multiple harvesters to be integrated, engineers should examine all possible energy sources. Taking a dual-source approach also allows, for example, PV to be the dominant source in summer, with thermos-electric generators in the winter when the heating is on.
While these calculations are widely available online, we find the best (certainly the most-simple) way to run this is via power profiling tool, such as those from Nordic or Silicon Labs.
From these calculations, not only can the ideal energy source / harvesting technology be chosen, but also the ideal power storage (be it a chemical battery or a supercapacitor) too. Here, it should be noted that not all combinations of storage element options can be used with every energy harvesters architecture: with a key criterion being to ensure the voltage ranges of each do not overlap. Indeed, a margin in the region of 0.3 V should be set between the voltage ranges of each element to ensure charging can be initiated regardless of the highly variable harvesting conditions. This is extensively discussed in recent webinar (Watch the Webinar Replay: Ambient IoT from Storage Element Stand Point – E-peas)
For this, it’s critical to ensure the features embedded into the PMIC match the application’s requirements. For example, a cold start capability will allow a system to initiate in conditions with exceptionally low available energy levels. A deep sleep mode will also enable a minimal power draw during idle periods. Each PMIC from e-peas is tailored to meet the needs of variables including specific source, storage types (see figure 4, below).
Selecting the right PMIC will also ensure the optimal MPP tracking algorithm is applied to extract the maximum possible power from the energy source. For IoT applications, low power usage and processing requirements are critical and the PMIC should therefore implement either an adaptive Open Circuit Ratio method such as FOCV, or a constant tracking based on either voltage or impedance.
The choice of each change according to the system and type of harvester used, e-peas can help you determine the best option for your system.
With IoT systems being used across a wide range of applications, each with highly specific power supply requirements, it’s vital to determine the supply voltage and current range. For voltage, the minimum and maximum power supply voltage need to be understood, as well as an awareness of the voltages that enable a better performance / better efficiency. For current, the maximum peak current the IoT device will draw needs to be known, as well as the idle current.
The final steps in implementing energy-harvesting to power an IoT device is to prototype and evaluate the system.
For this, many evaluation boards are available. For e-peas, this includes our own in-house developed boards, but there is also a fast-growing ecosystem and evaluation kits from Dracula Technologies, Silicon Labs are also available for prototyping and testing the system.
These kits will enable the testing of different MCUs, energy sources, storage technologies, PMICs, data transmission protocols and other on-board components and allow slimmer, more sustainable devices that stay ahead of both customer needs and of any incoming legislation.
For further information on implementing the steps listed in this blog, a more detailed white paper is available from www.e-peas.com.
