Ambient vibration is one of the most abundant yet most frustrating sources of renewable energy. It is everywhere—bridges sway, ocean waves roll, and human bodies move—but nearly all of it arrives at frequencies far below what conventional energy harvesters can capture. Now, a team of researchers led by Jianfeng Hong and Mingjie Guan of Xiamen University of Technology has unveiled a dual-beam bistable energy harvester that converts vibrations slower than one hertz into measurable electrical power, reporting a maximum output of 65.6 microwatts at just 1 Hz and an acceleration of 0.16 g. The work, published in Results in Engineering, represents a striking advance in the quest to power wireless sensor networks without batteries.
The fundamental problem the team confronted is a mismatch of scales. Piezoelectric energy harvesters, which generate charge when mechanically strained, are compact, mechanically simple, and efficient compared with electromagnetic or triboelectric alternatives. But their natural frequencies typically sit in the tens or hundreds of hertz, while environmental vibrations—from wave motion at roughly 0.1 Hz to bridge oscillations below 1 Hz and human-induced movement under 2 Hz—linger far lower. When excitation frequency falls well below resonance, a conventional harvester barely deforms, and its power output collapses. Bistable energy harvesters offer a clever escape: by engineering a potential energy landscape with two wells separated by a barrier, they can undergo dramatic snap-through oscillations between wells, producing large-amplitude motion even under weak, slow excitation.
The catch is that the potential barrier itself becomes an obstacle. If the barrier is too tall, the beam stays trapped in a single well and never achieves the high-energy inter-well orbit that generates useful voltage. Previous researchers have attacked this problem by adding more stability states—tri-stable, quad-stable, even penta-stable configurations—or by using asymmetric geometries and variable potential wells. These approaches lower the barrier but add regulatory complexity and can degrade adaptability under real-world low-frequency conditions. Dual-beam structures, in which two cantilevers are magnetically coupled, have emerged as a particularly promising route, though existing designs typically arrange the beams in opposite directions and demand considerable installation space.
The new device departs from that template in two significant ways. First, it fits inside a cylindrical tube, making it suitable for space-constrained deployments on bridges and similar infrastructure. Second, it incorporates a spring-magnet oscillator with a fixed pulley and rigid string, allowing a long spring to lie parallel to the beams. This pulley arrangement gives the oscillator a very low internal frequency, so it can respond to ultra-low-frequency ambient motion. The oscillator’s moving magnet, riding on a slider-guide rail, plucks the two piezoelectric cantilever beams—designated Beam A and Beam B—through nonlinear magnetic forces, realizing a frequency up-conversion mechanism: slow ambient vibration is transformed into fast beam oscillation near the beams’ natural frequencies, where piezoelectric conversion is most effective.
The physics underlying the device is captured in a detailed theoretical model. The three permanent magnets—two tip magnets at the beam ends and the moving magnet on the oscillator—are treated as magnetic dipoles, allowing the team to compute the magnetic potential energy as a function of beam and oscillator displacements. The resulting potential energy surface is asymmetric and saddle-shaped, with a self-decreasing barrier. When the oscillator sits at its equilibrium position, the potential barrier reaches its maximum; once external excitation pushes the oscillator, two distinct potential wells emerge for each beam, and the barrier drops, making snap-through inter-well motion far easier. Notably, the lower beam experiences a shallower barrier than the upper one, so it achieves larger snap-through amplitudes—a prediction the experiments would later confirm.
The model also revealed a critical design parameter: the magnetic distance between the beams and the oscillator. Numerical calculations showed that when this distance is below 12 millimeters, each beam possesses two equilibrium points flanking the zero-displacement position, forming an asymmetric bistable potential well. As the distance increases, the well shallows and the system transitions to a monostable state at exactly 12 millimeters. This theoretical threshold would prove decisive in the laboratory, where the researchers used a precision micro-stage to tune the magnetic distance and observe the system’s behavior directly.
The experimental campaign began with parametric studies at 0.7 Hz excitation with 4-centimeter amplitude. Five spring stiffnesses were tested, and at 13 N/m both beams reached their maximum peak-to-peak open-circuit voltages of 6.3 V and 6.1 V. Tip mass mattered as well: the lower beam peaked at 11.3 V with a 126-gram tip mass, while the upper beam reached 11 V at 136 grams. The magnetic distance experiments then confirmed the theoretical prediction with remarkable clarity. At 9 millimeters, the system remained trapped in a single well with weak output. At 12 millimeters, it crossed the barrier, entered the high-energy inter-well orbit, and produced its maximum voltage. Beyond that, at 15 millimeters, the response dissolved into chaotic oscillation, and with further separation the magnetic coupling weakened until the frequency up-conversion mechanism ceased entirely.
Frequency sweep tests from 0.5 to 1.0 Hz demonstrated the up-conversion mechanism in action. At 0.5 Hz, both beams responded sluggishly at the excitation frequency itself, indicating chaotic or intra-well motion. But at 0.7 Hz, the dominant spectral components jumped to 7.96 Hz and 8.12 Hz for the two beams—clear evidence that slow excitation had been converted into fast resonant oscillation. As frequency rose to 1.0 Hz, spectral amplitudes climbed to 4.75 V and 4.32 V. Under a constant acceleration of 0.16 g and sweeping load resistance from 50 to 400 kilo-ohms, the device delivered its headline performance: at 1 Hz and 120 kilo-ohms, Beam A produced 32.4 microwatts and Beam B 33.2 microwatts, for a combined 65.6 microwatts and a power density of 69.3 microwatts per cubic centimeter.
That figure places the new harvester at the top of its class. Compared against previously published ultra-low-frequency devices—including a sprung eccentric rotor producing 61.3 microwatts, a quin-stable harvester producing 48 microwatts, and an L-shaped cantilever producing 5.89 microwatts—the dual-beam design leads in both absolute power and power density, while operating at a notably gentle acceleration of 0.16 g. Head-to-head testing against an otherwise identical bistable harvester lacking the dual-beam structure showed the new design delivering 6.07 times the power and 3.78 times the voltage at 1 Hz, with phase portraits revealing the large-amplitude cyclic orbits responsible for the gain.
The authors point to several avenues for pushing performance further: bonding additional piezoelectric patches to each beam, rigidly fastening the tip magnets to the slider, and deploying Halbach magnet arrays to intensify the magnetic interaction. Future work will focus on miniaturizing and integrating the harvester for self-powered demonstrations in real settings, along with designing interface circuits for energy storage. For now, the device stands as a compelling proof of concept that the sub-hertz world—waves, bridges, slow human motion—need not remain an energy-harvesting dead zone. By combining a compact cylindrical form factor, a pulley-based low-frequency oscillator, and a self-shallowing bistable potential landscape, the dual-beam design transforms some of the slowest vibrations around us into a practical stream of electricity.
Subject of Research: A dual-beam bistable piezoelectric energy harvester for ultra-low-frequency vibration energy harvesting
Article Title: Design and experimental study of a novel dual-beam bistable energy harvester under ultra-low frequency vibration
Article References: Hong, J., Guan, M., Chen, J., Zhang, T., & Jin, Y. (2026). Design and experimental study of a novel dual-beam bistable energy harvester under ultra-low frequency vibration. Results in Engineering, 32, Article 113305. https://doi.org/10.1016/j.rineng.2026.113305
Image Credits: AI Generated
DOI: 10.1016/j.rineng.2026.113305
Keywords: energy harvesting, piezoelectric, bistable, ultra-low frequency, vibration, magnetic coupling, frequency up-conversion, potential barrier, wireless sensors, dual-beam, nonlinear dynamics, renewable energy
News Source: Denise Maddox. (October 6, 2026). Dual-beam energy harvester turns ultra-low-frequency vibrations into usable power. Scienmag.
