At The Things Conference 2025, a global hub for IoT innovation, we see energy harvesting playing a pivotal role in building autonomous devices that scale across industries. Solar energy harvesting, a key enabler of low-power embedded systems, is transforming how devices connect to LoRaWAN networks across industries. Together with Qoitech, Silicon Labs, and Mouser Electronics, we are showcasing how developers can leverage solar energy harvesting to unlock sustainable IoT deployments in real-world conditions.

This article breaks down the core architecture of an energy harvesting system and demonstrates how to evaluate real-world performance using the xG22E Explorer Kit and the Otii Ace Pro.

Energy harvesting enables the creation of maintenance-free embedded systems by converting ambient energy, such as light, into usable electrical power. These self-powered devices are ideal for applications where frequent battery replacements are impractical or costly, such as environmental sensors, asset trackers, and smart infrastructure.

However, designing an efficient energy harvesting system goes beyond connecting a device to a solar cell together with a storage element. Developers need to understand how much energy is harvested, how efficiently it flows through the system, and how the load consumes it over time. That means understanding the trade-offs in power path efficiency, maximum power point tracking, storage behavior, and system-level consumption.

What is an energy harvesting system?

An energy harvesting (EH) system, known as Ambient IoT across the Internet of Things ecosystem, converts ambient energy into electrical power and stores or uses it to operate a device. Depending on the application, it can harvest energy from light, motion, heat, or radio frequency. Enabling systems that can remain operational for years without battery replacements, as long as the harvested energy is sufficient to cover consumption.

Our focus in this article is on light-based energy harvesting, which is commonly used in indoor and outdoor IoT applications.

At a high level, an energy harvesting system must do three things:

1.Capture energy efficiently, even under varying light conditions.
2. Store energy with minimal losses, using rechargeable batteries or supercapacitors
3. Deliver energy reliably to the load, including during cold starts and peak demand.

While the concept is straightforward, implementing it in a real deployment involves trade-offs and limitations that make the system far more complex than it appears:

• Unpredictable energy availability
• Trade-offs between storage size, leakage, and recharge rates.
• Load profiles with high variability, such as radio transmissions, FOTA or sensor polling, which can drain storage quickly if not properly managed.
• Cost and complexity when transitioning from proof-of-concept to a manufacturable, scalable product.

Core components of a light-based harvesting system

An overview of a light-based energy harvesting architecture

Each component plays a critical role in ensuring that energy is captured, managed, and delivered efficiently from the environment to the device. Let’s break down the function and relevance of each one.

Photovoltaic (PV) cell
The PV cell converts ambient light into electrical energy through the photovoltaic effect. Choosing the proper cell requires considering the deployment environment, whether the device will operate outdoors, indoors under artificial lighting, or in a combination of both, as its performance depends on factors such as light intensity, spectral match, temperature, and the electrical load.

Power management IC (PMIC)
The PMIC manages how energy flows from the PV cell to the storage and the load. It performs DC-DC conversion (boost or buck) as needed, and often includes a Maximum Power Point Tracking (MPPT) algorithm to extract the most power possible under changing light conditions.

Energy storage
Harvested energy must be stored to cover periods when the light is insufficient or when the load draws more current than the PV cell can provide. Its selection will directly depend on current profile, energy availability, and system lifetime goals. Common options include:

• Rechargeable batteries, with high energy density and limited cycle life
• Supercapacitors, with fast response, long cycle life, and higher self-discharge

Load
The load is the embedded system consuming energy, typically composed of a microcontroller, sensors, radio modules, and supporting circuitry. Its behavior defines the system’s energy profile, which varies with operating states, sensor activity, and communication cycles. Proper load profiling is essential to match harvested energy with real-world consumption and ensure consistent system operation.

Key design considerations

To ensure the system remains operational under all varying conditions, developers must consider the following during evaluation:

• Power balance → Harvested energy must exceed average consumption. This balance is dynamic; it shifts with time of day, weather, and usage frequency. If the device transmits more frequently or enters low-light areas, the system may deplete its energy buffer and fail.
• PMIC (MPPT) efficiency → Since the maximum power point (MPP) varies with light intensity due to clouds, indoor lighting, or shading, accurate and responsive tracking is essential. A PMIC with adaptive MPPT can extract more power in partially lit conditions.
• Storage strategy → Storage should be selected based not only on capacity, but also on charge rate, leakage current, and cold-start behavior. For example, a supercapacitor may handle peak loads effectively but could leak energy too quickly in low-duty-cycle applications.
• Load profile → Measuring, understanding and profiling the device’s energy consumption over time is critical. This includes deep sleep current, wake-up time, sensor sampling, and radio transmission bursts. Even a small firmware change like an additional UART log, can introduce unexpected energy drain and lead to system failures in the field.

How to evaluate the design

The setup
A single, straightforward setup is enough to iterate on key design evaluations and prove whether your energy harvesting device maintains a net-positive.

Energy harvesting setup with Otii Ace Pro measuring PV cell and energy storage performance.

We used BRD8201A Dual Harvester Shield (designed with an e-peas AEM13920 PMIC) available in the xG22 Energy Harvesting Explorer Kit from Silicon Labs, an outdoor use Voltaic Systems PV cell, and a 10 F 3.8V lithium capacitor for storage.

To evaluate the system, we connected two Otii Ace Pro units in in-line (ampere meter) mode with 4-wire sensing. The first Otii Ace Pro measured energy harvested performance between the PV cell and the PMIC, while the second tracked storage charge dynamics between the PMIC and the capacitor. This configuration provided full visibility into energy flow from light input to storage, with enough resolution to capture critical behaviors such as MPPT tracking and cold-start handling.

The entire setup will also be featured on the Wall of Fame at The Things Conference 2025 and in booth B21, giving attendees an up-close look at how to validate low-power designs for scalable deployments.

A full breakdown of this setup, including measurement data and deeper insights, can be found in this dedicated study.

The setup is scalable: add another Otii Ace Pro to simulate the load. If you only need to evaluate the harvester, storage, and IoT device separately, a single Otii Ace Pro is sufficient.

Learnings from the evaluations

Using the Otii Ace Pro, you can monitor the direction of current flow in real time. In this example setup it revealed a critical insight: during short bursts of device activity, the storage element temporarily supplied energy to the system since the harvested energy was not enough. Immediately after, when the device returned to idle mode, the storage began to recharge, replenished by the PV cell.

The important finding here is that even though the energy storage experienced short-term depletion during activity, the overall balance across the full duty cycle remained sustainable. In other words, the device harvested more than it consumed, allowing it to remain energy net-positive.

Example of PV and energy storage performance evaluation: illumination level effects.

However, understanding this short-term balance is not enough. To design for reliability, it is essential to evaluate the system over longer periods and under varying light conditions. Indoor applications, for instance, present unique challenges, as light levels can fluctuate dramatically depending on the time of day, occupancy, or even extended holiday periods when artificial lighting may be turned off altogether. PV cells also age and lose in their efficiency. Without careful long-term evaluation, a device that appears self-sufficient under steady lab lighting might fail in a real-world deployment. Hence, look to understand both short-term balance and long-term balance to conclude energy net-positive.

Another key factor is the aging of the storage component. Over time, capacitors and batteries degrade, losing both capacity and efficiency. By emulating both new and aged profiles with the Otii Ace Pro’s battery emulator mode, developers can anticipate how their system will perform years down the line, rather than just at the initial deployment. This approach provides valuable foresight, helping to avoid unpleasant surprises when storage elements no longer deliver the expected support during low-light periods.

Together, these findings underline the importance of looking beyond instantaneous measurements. A robust evaluation of energy harvesting systems requires careful analysis of storage behavior, real-time energy balance, long-term resilience, and the inevitable effects of component aging. Only by combining these perspectives can engineers build light-powered IoT devices that are not only functional in the lab but also reliable and self-sustaining in real-world conditions.

Conclusion

Energy harvesting offers a clear path to building truly autonomous embedded systems, but only when backed by careful design, measurement, and validation. By leveraging platforms like the xG22E Energy Harvesting Explorer Kit and the Otii Ace Pro, developers can fine-tune their devices for efficiency and reliability. To make getting started easier, Mouser has put together all the hardware components into a ready-to-use kit, so you can focus on validation and development.

At this year’s The Things Conference, spot the EFR32xG22E Energy Harvesting Explorer Kit on the Wall of Fame, giving you the chance to explore how light-powered innovation is shaping the future of low-power, long-range connected solutions. Want to see the evaluation set-up live? Join the global IoT community at The Things Conference 2025 alongside the teams at Qoitech**, Silicon Labs and Mouser Electronics** and learn how to integrate and validate light-based energy harvesting in your next project.

If you haven't got your tickets yet, join us in Amsterdam for The Thing Conference. Don't miss out IoT's biggest gathering.