Powering Autonomous Sensors Without Batteries

Wireless sensors are everywhere: on bridges and pipelines, in buildings, on farms and inside machines. Most of them run on primary batteries, and every battery eventually needs to be changed. For one sensor that is a nuisance. For a network of thousands, spread across a factory or a city, it becomes the dominant operating cost, and every missed replacement is a gap in the data. Batteries also limit where sensors can go, because a node that cannot be reached cannot be serviced. The obvious answer is to power the sensor from its surroundings. The difficulty is that ambient sources are weak and intermittent, while a sensor needs a dependable supply at the exact moment it measures and transmits. Matching the two takes careful accounting of power, time and storage.

The Established Approaches

Battery-free and battery-extending sensors are an active field with working products. Four elements recur.

Low-power design and duty cycling. A typical node sleeps most of the time, wakes briefly to measure, and transmits in short bursts. Modern microcontrollers sleep at a few microamperes, and low-power radios such as Bluetooth Low Energy or LoRa transmit for milliseconds. This brings the average power of many nodes into the range of 10 to 100 µW [1, 2].

Single-source harvesters. Indoor photovoltaic cells power room sensors and calculators; thermoelectric generators run sensors on warm pipes and machines; vibration harvesters work on motors and compressors; radio-frequency harvesting powers RFID-based sensing tags at short range [2, 3, 4]. Each works where its source is reliable.

Storage as buffer. Supercapacitors or small rechargeable cells absorb the energy between bursts, so the harvester only has to cover the average load, not the peak [2].

Energy-neutral operation. Power management adapts the duty cycle to the energy available, so the node never spends more than it collects over a full cycle [5].

The limits are well known. Indoor light switches off at night. A temperature difference can disappear when a process stops. Vibration harvesters are tuned to one frequency. RF harvesting depends on the distance to a transmitter. A single-source node is therefore sized for the worst period of its one source, which makes it larger and more expensive than its average consumption would suggest.

The Open Question

Every established design relies on one dominant source and sizes storage around its gaps. What is not well described is a material that couples several weak ambient inputs at once: heat, electromagnetic fields, vibration and particle interactions, all acting on the same layers at the same time. If such inputs are collected together, the gaps of one source are partly filled by the others, and the supply becomes steadier. To size a sensor on that basis, the engineer needs one model that adds all coupled inputs inside the material and keeps conservation of energy explicit.

A Generalized Approach

A generalized approach to this problem is represented by the Schubart Master Formula, which gives the usable electrical output of a material under several ambient inputs:

P(t) = η · ∫_V Φ_eff(r, t) · σ_eff(E) dV

The derivation, notation and open questions are published at pt-schubart.com. The model is introduced in Mathematical Models for Multi-Source Energy Harvesting.

Definitions

P(t): instantaneous electrical output, the usable electrical power delivered at time t. Unit: watt (W). Energy over an interval is the time integral of P(t), in joules or watt-hours.
η: total conversion efficiency, dimensionless, between 0 and 1. It combines coupling, resonance, rectification and electrical losses.
Φ_eff(r, t): effective ambient flux density at position r and time t, the local strength of all ambient inputs acting on the material.
σ_eff(E): energy-dependent effective interaction cross-section, how strongly the material couples to an input of energy E.
∫_V: volume integral over the active material.

Together, Φ_eff · σ_eff is the coupled power per unit volume (W/m³), so the integral over V yields watts. How that product divides between the two factors depends on the input channel, so no separate units are assigned to them here.

Assumptions and Boundaries

The Schubart Master Formula is a proposed generalized framework, not an established equation of physics. Its assumptions and limits:

The material is an open, non-equilibrium system that exchanges energy with its surroundings.
Output is bounded by coupled input power: P_out ≤ ΣP_in.
Coupled power densities and η must be measured on the specific material and at the specific site.
A sensor’s supply must cover its average load over a full cycle; storage covers the bursts.
Whether a module covers part of the load or all of it depends on consumption and operating profile. A sensor module is an integration stage, not a finished self-powered product.

Worked Example

All values below are illustrative values, chosen to show the method. They are not measured results.

The load. A sensor node runs at 3 V:

Mode	Current	Duration	Average power
Sleep	2 µA	always	6.0 µW
Measurement	1 mA	10 ms every 60 s	0.5 µW
Transmission	15 mA	20 ms every 60 s	15.0 µW
Total			21.5 µW

Battery only. A coin cell storing about 0.675 Wh lasts 0.675 Wh ÷ 21.5 µW ≈ 31,400 hours, about 3.6 years.

Small module. An active layer of 1 cm × 1 cm with a thickness of 0.1 mm has V = 1 × 10⁻⁸ m³. With a combined coupled power density Φ_eff · σ_eff of 2,800 W/m³ from thermal, electromagnetic and mechanical inputs, ΣP_in = 28 µW. At η = 0.2, P = 5.6 µW, and the boundary holds (5.6 µW ≤ 28 µW). The module covers about a quarter of the load continuously, and the same coin cell lasts 0.675 Wh ÷ 15.9 µW ≈ 42,500 hours, about 4.8 years.

Larger module. With four times the area, P = 22.4 µW, slightly above the average load. The primary battery could then be replaced by a small supercapacitor that only bridges the transmission bursts, depending on consumption and operating profile.

The example shows the two design levers that matter most for sensors: the average load, set by the duty cycle, and the active volume of the harvesting layer.

Where It Leads

At the sensor scale, the output of a few square centimetres of material decides whether batteries are changed less often or not at all. The material itself, layered graphene and semiconductor structures, sets the coupling. The step from sensors to portable devices, appliances and building-scale units is described in From Sensor to Grid, and the general calculation method in How to Calculate Ambient Energy Harvesting.

References

J. A. Paradiso and T. Starner, “Energy scavenging for mobile and wireless electronics,” IEEE Pervasive Computing 4(1), 18–27 (2005).
R. J. M. Vullers, R. van Schaijk, I. Doms, C. Van Hoof and R. Mertens, “Micropower energy harvesting,” Solid-State Electronics 53, 684–693 (2009).
S. Roundy, P. K. Wright and J. Rabaey, “A study of low level vibrations as a power source for wireless sensor nodes,” Computer Communications 26, 1131–1144 (2003).
A. P. Sample, D. J. Yeager, P. S. Powledge, A. V. Mamishev and J. R. Smith, “Design of an RFID-based battery-free programmable sensing platform,” IEEE Transactions on Instrumentation and Measurement 57, 2608–2615 (2008).
A. Kansal, J. Hsu, S. Zahedi and M. B. Srivastava, “Power management in energy harvesting sensor networks,” ACM Transactions on Embedded Computing Systems 6(4), article 32 (2007).
H. J. Visser and R. J. M. Vullers, “RF energy harvesting and transport for wireless sensor network applications: principles and requirements,” Proceedings of the IEEE 101, 1410–1423 (2013).
Schubart Master Formula: derivation, notation and open questions, pt-schubart.com.

This page is published by the Neutrino Engineering District. Back to Fundamentals